Power Quality in Power Distribution Systems: Concepts and Applications [1 ed.] 0367750910, 9780367750916

The comprehensive textbook will help readers to develop analytic reasoning of power quality aspects in distribution powe

320 96 17MB

English Pages 372 Year 2023

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Power Quality in Power Distribution Systems: Concepts and Applications [1 ed.]
 0367750910, 9780367750916

Citation preview

Power Quality in Power Distribution Systems The comprehensive textbook will help readers to develop analytic reasoning of power quality aspects in distribution power systems. It will as an ideal study material for senior undergraduate and graduate students in the field of electrical engineering, electronics and communications engineering. • Provides explanation of transformations and power theories for single phase and three-phase systems. • Discusses concepts illustrating power quality aspects in power distribution network. • Examines detailed derivations and analysis of voltage and current compensation techniques. •  Discusses custom power devices such as DSTATCOM, DVR and UPQC. •  Presents solved examples, theoretical and numerical exercises in each chapter. This textbook comprehensively covers fundamentals concepts of power quality with the help of solved problems. It provides basic understanding of power quality aspects in power systems, especially in power distribution networks and explains issues related to power quality problems, their quantification, analysis and interpretation. It covers important topics including single phase circuits, three phase circuits, theory of fundamental load compensation, instantaneous reactive power theory, theory of instantaneous symmetrical components, dynamic voltage restorer (DVR) and unified power quality conditioner. Pedagogical features including solved problems and unsolved exercises are interspersed throughout the text for better understanding.The textbook is primarily written for senior undergraduate and graduate students in the field of electrical engineering, electronics and communications engineering for courses on power quality/power system/ power electronics. The textbook will be accompanied by teaching resource including solution manual for the instructors.

Power Quality in Power Distribution Systems Concepts and Applications

By Mahesh Kumar Mishra

First edition published 2024 by CRC Press 2385 NW Executive Center Drive, Suite 320, Boca Raton FL 33431 and by CRC Press 4 Park Square, Milton Park, Abingdon, Oxon, OX14 4RN CRC Press is an imprint of Taylor & Francis Group, LLC © 2024 Mahesh Kumar Mishra Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers have attempted to trace the copyright holders of all material reproduced in this publication and apologize to copyright holders if permission to publish in this form has not been obtained. If any copyright material has not been acknowledged please write and let us know so we may rectify in any future reprint. Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying, microfilming, and recording, or in any information storage or retrieval system, without written permission from the publishers. For permission to photocopy or use material electronically from this work, access www.copyright.com or contact the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. For works that are not available on CCC please contact [email protected] Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for identification and explanation without intent to infringe. ISBN: 978-0-367-75091-6 (hbk) ISBN: 978-1-032-61729-9 (pbk) ISBN: 978-1-032-61730-5 (ebk) DOI: 10.1201/9781032617305 Typeset in Nimbus font by KnowledgeWorks Global Ltd.

Dedicated to my parents: Late Shri Gendan Lal Mishra and Late Smt. Ramkali Mishra

Contents About the Author ....................................................................................................xiii Foreword .................................................................................................................. xv Preface....................................................................................................................xvii Chapter 1

Introduction to Power Quality in Power Distribution System......... 1 1.1 1.2 1.3

Introduction............................................................................. 1 Power Quality ......................................................................... 2 Power Quality Problems ......................................................... 4 1.3.1 Transients ................................................................... 6 1.3.2 Short-Duration Variations .......................................... 8 1.3.3 Long-Duration Variations .......................................... 9 1.3.4 Imbalance................................................................. 11 1.3.5 Waveform Distortion................................................ 11 1.3.6 Voltage Fluctuations................................................. 13 1.3.7 Power Frequency Variations .................................... 13 1.4 Mitigation Techniques .......................................................... 14 1.4.1 Network Re-configuring Type ................................. 14 1.4.2 Compensating Type ................................................. 17 1.5 Power Quality Monitoring .................................................... 19 1.6 Power Quality Indices........................................................... 20 1.7 Power Quality Standards ...................................................... 24 1.8 CBEMA and ITIC Curves .................................................... 26 1.9 Summary............................................................................... 27 1.10 Problems ............................................................................... 27 References...................................................................................... 28 Chapter 2

Single-Phase Circuits: Power Definitions and Components.......... 31 2.1 2.2 2.3 2.4 2.5

Introduction........................................................................... 31 Power Terms and Definitions in Single-Phase Systems ....... 31 Phasor Representation of Electrical Quantities .................... 35 Sinusoidal Voltage Source Supplying Non-linear Load Current .................................................................................. 41 Non-sinusoidal Voltage Source Supplying Linear and Non-linear Loads .................................................................. 48

vii

viii

Contents

2.5.1 Active Power............................................................ 50 2.5.2 Reactive Power......................................................... 51 2.5.3 Apparent Power ....................................................... 52 2.5.4 Non Active Power .................................................... 54 2.5.5 Distortion Power ...................................................... 54 2.5.6 Fundamental Power Factor ...................................... 54 2.5.7 Power Factor ............................................................ 55 2.6 Summary............................................................................... 62 2.7 Problems ............................................................................... 62 References...................................................................................... 69 Chapter 3

Three Phase Circuits: Power Definitions and Various Components................................................................................... 71 3.1 3.2

Introduction........................................................................... 71 Three-phase Balanced System.............................................. 72 3.2.1 Three-phase Instantaneous Power............................ 72 3.2.2 Three Phase Instantaneous Reactive Power............. 74 3.2.3 Power Invariance in abc and αβ 0 Coordinates ....... 76 3.3 Instantaneous Active and Reactive Powers for Three-phase Circuits............................................................. 78 3.3.1 Three-Phase Balance System................................... 78 3.3.2 Three-Phase Unbalanced System............................. 80 3.4 Symmetrical Components..................................................... 83 3.4.1 Power Invariance in Symmetrical Components Transformation, Vector and Arithmetic Apparant Powers ...................................................... 89 3.4.2 Effective Apparent Power ........................................ 94 3.4.3 Positive Sequence and Unbalance Powers............... 95 3.5 Three-phase Balanced Nonsinusoidal System.................... 107 3.5.1 Neutral Current ...................................................... 108 3.5.2 Line to Line Voltage............................................... 108 3.5.3 Apparent Power with Budeanu Resolution: Balanced Distortion Case....................................... 110 3.5.4 Effective Apparent Power for Balanced Non-sinusoidal System .......................................... 110 3.6 Three-phase Unbalanced and Non-sinusoidal System ....... 115 3.6.1 Arithmetic and Vector Apparent Power with Budeanu’s Resolution ............................................ 118 3.6.2 Effective Apparent Power ...................................... 119 3.7 Summary............................................................................. 127 3.8 Problems ............................................................................. 127 References.................................................................................... 135

ix

Contents

Chapter 4

Theory of Fundamental Load Compensation .............................. 137 4.1 4.2

Introduction......................................................................... 137 Theory of Fundamental Load Compensation ..................... 137 4.2.1 Power Factor Correction ........................................ 138 4.2.2 Voltage Regulation................................................. 140 4.2.3 An Approximation Expression for the Voltage Regulation .............................................................. 143 4.3 Some Practical Aspects of Compensator used as Voltage Regulator ............................................................... 151 4.4 Phase Balancing and Power Factor Correction of Unbalanced Loads .............................................................. 154 4.4.1 Three-phase Unbalanced Loads............................. 154 4.4.2 Representation of Three-phase Delta Connected Unbalanced Load ................................................... 157 4.4.3 An Alternate Approach to Determine Phase Currents and Powers .............................................. 160 4.4.4 An Example of Balancing an Unbalanced Delta Connected Load ..................................................... 162 4.5 A Generalized Approach for Load Compensation using Symmetrical Components................................................... 164 4.5.1 Sampling Method................................................... 172 4.5.2 Averaging Method.................................................. 173 4.6 Compensator Admittance Represented as Positive and Negative Sequence Admittance Network ........................... 175 4.7 Compensation of Star Connected System with Grounded Neutral ................................................................................ 188 4.8 Summary............................................................................. 198 4.9 Problems ............................................................................. 198 References.................................................................................... 202 Chapter 5

Control Theories for Load Compensation................................... 204 5.1 5.2

5.3 5.4

Introduction......................................................................... 204 Instantaneous Reactive Power Theory................................ 205 5.2.1 State Space Modeling of the Compensator............ 217 5.2.2 Switching Control of the VSI ................................ 219 5.2.3 Generation of Ploss to Maintain dc Capacitor Voltage ................................................................... 220 5.2.4 Computation of Load Average Power.................... 221 Some Misconceptions in Instantaneous Reactive Power Theory................................................................................. 222 Theory of Instantaneous Symmetrical Components........... 242 5.4.1 Compensating Star Connected Load...................... 242 5.4.2 Compensating Delta Connected Load ................... 251

x

Contents

5.4.3

Compensation for Three-phase Three-wire System Supplying a Passive Load ......................... 257 5.4.4 Equivalence Between the Passive and Active Load Compensation for Three-phase Three-wire System ................................................. 258 5.5 Summary.............................................................................269 5.6 Problems .............................................................................270 References.................................................................................... 275 Chapter 6

Voltage Compensation Using Dynamic Voltage Restorer ...........277 6.1 6.2 6.3 6.4

Introduction......................................................................... 277 Methods to Regulate Voltage.............................................. 278 Dynamic Voltage Restorer (DVR) ...................................... 278 Operating Principle of DVR ............................................... 280 6.4.1 General Case ..........................................................282 6.5 Mathematical Description to Compute DVR Voltage ........ 285 6.6 Transient Operation of the DVR......................................... 289 6.6.1 Operation of Three Phase DVR with Unbalance Voltages without Harmonics ..................................290 6.6.2 Operation of Three Phase DVR with Unbalance Voltages with Harmonics ....................................... 291 6.7 Realization of DVR voltage using Voltage Source Inverter................................................................................292 6.8 Maximum Compensation Capacity of the DVR without Real Power Support from the dc Link ................................ 295 6.9 Summary.............................................................................310 6.10 Problems .............................................................................310 References.................................................................................... 314 Chapter 7

Unified Power Quality Conditioner............................................. 317 7.1 7.2 7.3 7.4

7.5

7.6

Introduction......................................................................... 317 UPQC Structure .................................................................. 318 Operation and Control of UPQC ........................................ 322 Extraction of Reference Currents for Shunt VSI using Instantaneous Symmetrical Component Theory................. 323 7.4.1 Generation of Reference Currents for Shunt VSI ......................................................................... 323 7.4.2 Control of Shunt Voltage Source Inverter ..............325 Control of Series Voltage Series Inverter............................325 7.5.1 Type-1 Control for Reference Voltage Generation of Series VSI ....................................... 326 7.5.2 Type-2 Control for Reference Voltage Generation of Series VSI ....................................... 330 Summary.............................................................................347

Contents

xi

7.7 Problems .............................................................................348 References.................................................................................... 352 Appendix................................................................................................................ 355 Appendix A

Fundamental and Positive Sequence Extraction.......................... 357 A.1 Fundamental Extraction......................................................357 A.2 Extraction of Positive Sequence Fundamental Components ........................................................................ 358

Appendix B

Answers to Numerical Problems ................................................. 361

Index ...................................................................................................................... 369

About the Author Prof. Mahesh Kumar Mishra received the B.Tech. degree from the College of Technology, Pantnagar, India, in 1991, the M.E. degree from the Indian Institute of Technology, Roorkee, India, in 1993, and the Ph.D. degree from the Indian Institute of Technology, Kanpur, India, in 2002, all in Electrical Engineering. He has 30 years of teaching and research experience. For about ten years (1993–2003), he was with the Department of Electrical Engineering, Visvesvaraya National Institute of Technology, Nagpur, India. Prof. Mahesh has been with the Indian Institute of Technology Madras since 2003 and is currently a Professor in the Department of Electrical Engineering. His research interests include the areas of power quality, power distribution systems, power electronic applications in power systems, microgrids, and renewable energy systems. Prof. Mahesh is a Life Member of the Indian Society of Technical Education and received IETE Prof. Bimal Bose Award for his outstanding contributions to Power Electronics Applications in Power Systems in 2015. He is a Fellow of the Indian National Academy of Engineering (FNAE) and the Institute of Engineers (India). He has completed a dozen of sponsored projects and consultancies. Under his research supervision, 18 Ph.D. and 12 M.S. have been awarded. Prof. Mahesh and his research scholars have been conferred with many International, National, and Institute level awards. He has 250 research publications in International and National peer-reviewed journals and conferences and delivered numerous expert talks in the areas of power quality, distributed generation, and microgrid systems. In his leisure, Prof. Mahesh loves to read and write on diverse subjects relating to people, life, culture, the environment, and the world. He is passionate about making music, tunes, and rhythms, and writing songs.

xiii

Foreword Power Quality and Reliability are the two most crucial aspects for customers connected to power distribution networks. While customers would like to have continuity of power supply, they would also like to distortion-free, near sinusoidal voltages at their supply inlets of specified voltage magnitude and frequency. With the increased penetration of converter-interfaced renewable energy resources like rooftop photovoltaics and batteries, the power distribution systems are facing many power quality challenges, such as voltage rise, harmonic distortion, phase imbalance, etc. Therefore, the study of power quality problems are their mitigation techniques are becoming extremely crucial for network planners and practicing engineers. This book is a welcome addition to the literature as it covers all the basics of power quality problems and their mitigation techniques. The author of the book, Professor Mahesh Kumar Mishra, has over 25 years of research experience in related areas. He has developed the theories of some of the mitigation techniques, developed laboratory prototypes, and implemented converterbased power conditioning devices. His encyclopedic knowledge in the related areas has resulted in this book. The topics are presented in a systematic manner, and the most important topics are covered in depth. The book is organized into seven chapters. Brief descriptions of the chapters are given below.  Chapter 1 introduces the book, where different power quality problems in power distribution systems, power quality mitigation techniques using power electronic-based devices, power quality indices, standards, and power acceptability curves are discussed.  Chapter 2 presents analyses of single-phase circuits, both in the presence and absence of harmonic distortions. Some important concepts like displacement power factor and active and reactive power in the presence of harmonic distortions are presented.  Analyses of three-phase circuits are presented in Chapter 3, where both balanced and unbalanced circuits are considered. Some important concepts like instantaneous real and reactive power, and apparent power for distorted and unbalanced circuits are discussed.  The fundamentals of load compensation are discussed in Chapter 4, where the aspects of voltage regulation, power factor correction, and load balancing for both delta and star-connected loads are discussed.  Converter-based load compensation, along with converter control is discussed in Chapter 5, where the load compensation using both the instantaneous reactive power theory and the theory of instantaneous symmetrical components are discussed. xv

xvi

Foreword

 Chapter 6 discusses the dynamic voltage regulator (DVR) and its operating principles for both balanced sinusoidal and unbalanced non-sinusoidal cases. The realization of a DVR through a power converter with associated results is also presented.  Chapter 7 presents unified power quality conditioner (UPQC), where both left shunt and right shunt configurations are discussed. This chapter also includes reference current and voltage generation and control of shunt current and series voltage through back-to-back connected voltage source converters. All the mathematical derivations of the underlying concepts are presented in detail. Furthermore, numerous worked-out examples are presented to aid and enhance learning and problem sets are provided at the end of each chapter that are useful for the students and their instructors. The depth and breadth of coverage of the book make this an exceptionally valuable addition to power quality studies that will help senior undergraduate and postgraduate students, instructors, and practicing engineers. Arindam Ghosh Professor, School of Electrical Engineering Computer and Mathematical Sciences Curtin University, Australia

Preface In last few decades, there is growing awareness of power quality in power industries and residential usages. There is more concern about reducing the electricity bill for its given usage and at the same time there is demand for high quality and reliability of power supply. To meet these criteria in power system operation, a power system engineer should be well aware of power quality concepts, terms, definitions, methods to enhance power quality, mitigation techniques, and devices. This book is an attempt to provide a comprehensive understanding of the subject matter in a simple way. The book aims to provide a basic understanding of power quality aspects in power systems, especially in power distribution networks. It brings clarity on various issues related to power quality problems, their quantification, analysis, and interpretation. In this course, students will develop skills to analyze power quality aspects in power systems and provide appropriate solutions using custom power devices. First, the concepts are presented from the basic principles, and then insightful expressions have been derived, followed by examples. At the end of every chapter, there is a sufficient number of questions to strengthen the knowledge. The book acts as an interface between the conceptual understanding of the subject and advanced research. After going through the book and studying it thoroughly, the student will have the confidence to continue further research in this area. All concepts discussed in the book are explained and analytic expressions are derived from the basic principles, without assuming any expressions or formulae from other references. This makes easy and direct understanding of the concepts and their applications. The insights from the derived expressions are discussed and elaborated, which encourages students to think analytically. Although there are many books on power quality, they are not developed as text books. Here is an attempt to understand power quality in a simple and lucrative way to serve as text book for undergraduate and graduate students in electric power areas at various technical universities and institutions. The book aims to develop a thought process and strong analytic reasoning of power quality aspects in power systems. The book has the following salient features.  Insightful and clear understanding of various terms, definitions, transformations, and power theories for single-phase and three-phase systems.  Lucid and clear illustration of power quality aspects in power distribution network with examples.  Detailed derivations and analysis of voltage and current compensation techniques.  Detailed explanation of custom power devices such as reactive network compensators, DSTATCOM, DVR, UPQC. xvii

xviii

Preface

 Theoretical and numerical exercises to practice concepts described in each chapter. The book is written in seven chapters. Chapter 1 focuses on various aspects of power quality, such as definitions of various terms in power quality, nature of power quality mitigation devices, monitoring of power quality, and standards in power quality. Chapter 2 aims to understand various power terms relating to power quality in a single-phase system with sinusoidal and non-sinusoidal voltage and current. Threephase systems, which are widely used in industries due to their high quality, reliability, and efficiency, are explored in Chapter 3. It focuses on understanding the behavior of the three-phase system under different conditions of voltages and currents. While analyzing three-phase circuits, αβ 0 transformation, and instantaneous symmetrical components transformation are derived and explained from the basic principles. Various types of apparent powers and the corresponding power factors are explained with practical applications, such as the design of distribution lines or feeders. After having a good understanding of power quality aspects, the focus is moved toward mitigation techniques. Chapter 4 discusses about voltage regulation and fundamental load compensation using purely reactive networks. The reader will learn from the discussed concepts and examples, how a delta-connected load or threephase ungrounded system can be fully compensated using purely reactive elements such as inductance and capacitance. The reader’s inquisitiveness is arisen by a simple question, whether these methods will work for three-phase grounded system? The answer to this leads to exploring active power filters as compensators. Thus, as a natural progression of flow, Chapters 5, 6, and 7 discuss methods employing active power filters working as Distribution Static Compensator (DSTATCOM), Dynamic Voltage Restorer (DVR), and Unified Power Quality Conditioner (UPQC), respectively, followed by the Appendix. The content of each chapter is summarized, followed by set of problems and relevant references. The problems given at the end of each chapter are thought stimulating and emphasize a deeper understanding of the concepts described. The keys to numerical problems are provided at the end of the book to verify the answers for the convenience of the readers. The book’s content, structure, and flow make it ideally suitable as a textbook for “Power Quality in Distribution Networks: Concepts and Applications” in many universities and institutions worldwide. Additionally, it can also be used as a reference book on power quality.

ACKNOWLEDGMENTS First and foremost, I would like to thank my department and the institute for rendering their support in writing the book. I thank all faculty colleagues in the Department of Electrical Engineering, Indian Institute of Technology Madras, Chennai. I am especially thankful to my colleagues Prof. R. Sarathi, Prof. B. Kalyan Kumar, Prof. S. Krishna, Prof. K. S. Swarup, Prof. Lakshmi Narasamma, Prof. S. Srinivas, Prof. Arun Karuppaswami, Prof. Krishna Vasudevan, and Prof. Kamalesh Hatua for their

Preface

xix

lively discussion on various technical aspects of the subject area, which led to clarity and fluidity in presenting the concepts. Their frank and freewheeling discussions during meetings and tea time have been thought-provoking, thus serving food for mind, body, and soul. I thank Prof. Srikanthan Sridharan, Engineering Design Department, IIT Madras for proofreading the book and inspiring thoughtful discussions in the subject area as well as in life matters. I have no words to express my heartfelt gratitude to Prof. Arindam Ghosh and Prof. Avinash Joshi, who motivated me on the path of exploration in the form of research and teaching during my Ph.D. days at IIT Kanpur. Without their light of knowledge, I would not have seen the shiner, brighter, and more affluent side of my professional as well as subjective worlds. I am thankful to Prof. S. C. Srivastava and Prof. Santanu Mishra at IIT Kanpur for their long association, encouragement, and inspiration. For about ten years (1993–2003), I was faculty at the Department of Electrical Engineering, Visvesvaraya National Institute of Technology (VNIT), Nagpur. I continue to have strong association and interaction with the institute and convey thanks to Prof. H. M. Suryawanshi, Prof. M. S. Ballal, and Prof. D. R. Tutakne for many interactive technical sessions through our research collaborations, as well as philosophical discussions. Prof. Yashwant Katpatal at VNIT Nagpur has recorded his friendship in my heart and has always been supportive as my own family. I would like to thank all my current research students, namely, Rajarshi Basu, Rohan Madnani, Nakka Pruthvi Chaithanya, Lokesh N, Hariharan R., Durga Malleswara Rao, Nafih Mohammad, Tony Thomas, Nimitha Muraleedharan, Ajit Upadhiya, Abhisek Panda, Leelavathi for their contributions to the book, by preparing circuit diagrams, verification of solutions, proofreading of the chapters and many other tasks involved. My graduated Ph.D. and M.S. scholars, D. K. Karthikeyan, Dr. G. Vincent, Dr. Koteswara U., Dr. S. Sasitharan, Dr. Srinivas B. Karanki, Dr. Siva K. Ganjikunta, Dr. Chandan Kumar, Dr. Sathish Kollimalla, Dr. Narsa Reddy Tummuru, Dr. Nagesh Geddada, Dr. Manoj Kumar M.V., Dr. Sijo Augustine, Dr. T. Sreekanth, Dr. Jakeer Hussain, Dr. Srikanth Kotra, Dr. R. Satish, Dr. S. Srikanthan, Dr. J. Suma, Dr. Nikhil Korada, Dr. P. Harshvardhan, Dr. Linash P. K., Manik Pradhan, V. Leela Krishna, Y.A.P. Ramshankar, Jaganath K., N. Karthikeyan, Anil Ramakuru, and K. Sridhar, deserve their names to be mentioned for their exemplary research works, which directly or indirectly have been a great help to write the book. I thank Mr. Gauravjeet Singh Reen, Senior Editor-Engineering, CRC Press, and his team members for all their support to make this book in publication form. My childhood passed in a small village Chausara, in the northern part of the country, surrounded by a beautiful canal and a river, enriched with greenery and agricultural land. It is really difficult to forget childhood days with a large family, relatives, and friends with all golden memories, good or bad. I thank them all from the bottom of my heart for the rich and diverse experiences of my life, which provided me with a subtle sense of music, an intuitive understanding of things, and feelings to know about people, plants, birds, animals, and inanimate beings, endowing the capacity to see the most primitive to the most evolved with the purity of connection among

xx

Preface

them. For all this journey of life, I am indebted to my parents, late Shri Gendan Lal Mishra and late Smt. Ramkali Mishra for their love and moral support in all my endeavors. I am grateful to my elder brother late Shri Jugal Kishore Mishra, younger brother Dr. Sarvesh Kumar Mishra, and sister Smt. Kumkum Sharma and their families for their love, support, and inspiration in all situations of my life. I also thank my late father-in-law Shri Chhote Lal Sharma, and mother-in-law, Smt. Bina Sharma for their unconditional love, inspiration, and motivation. I thank my wife, Kumud Mishra, whose inspiration always shows the way to remain focused. I am thankful to my daughter Niharika Mishra and son Tanish Kumar Mishra, who have been a constant source of enthusiasm and inspiration to complete the venture of writing this book. Last but not least, I am thankful to all known and unknown people, things, and factors who directly or indirectly have led to this creation. Mahesh Kumar Mishra Professor, Department of Electrical Engineering Indian Institute of Technology Madras Chennai, India

to Power 1 Introduction Quality in Power Distribution System

1.1

INTRODUCTION

The evolution of electric power systems has taken over one and a half centuries, and since then it has become an essential aspect of our lives. It started with the development of the dc power system in 1881. The dc power system was simple to realize and dealt with real quantities such as voltage, current, and resistance. However, the power could not be transmitted efficiently at higher voltages, which is required to minimize power losses over long-distance transmission. Around the same period, transformers and three-phase induction machines were developed, which laid the foundation for ac power system all over the world. Throughout the first half of the 19th century, with the advent of synchronous generators, ac power system was in full development with efficient power transfer over long distances at high voltage using step-up transformers. Later on, many such generating units were pooled together using transmission lines to form the ac grid as we know it today. This allowed bulk power transmission over long distances and resulted in more efficient and flexible operation of power system. Initially, the main objective of the power system was to deliver power to everyone and connect households. But in the late 1900s, with significant development of the electronics industry, the use of power-sensitive electronics devices started increasing rapidly in all sectors of the economy, such as daily household, industry, commerce, business, trade, finance, and healthcare. However, these sensitive devices and products required a clean and reliable supply of power. Subsequently, industrial growth had also started to increase rapidly, which led to increased use of advanced industrial machines and equipment that also needed a clean and reliable supply of electric power. This technological change led to growing concerns for electric utilities and end-users about the quality of electric power. The modern ac power system is quite flexible in terms of voltage an current levels. But at the same time, it has many challenging issues such as unbalance, harmonics, blackout, brownout, steady state and transient stability, reactive power, harmonics power, etc. Over the years, with extensive research in this domain, power engineers developed the existing power system, which allowed us to monitor and control the power system parameters to maintain reliable and good quality of power at all the power system levels, generation, transmission, and distribution.

DOI: 10.1201/9781032617305-1

1

2

1.2

Power Quality in Power Distribution Systems: Concepts and Applications

POWER QUALITY

Power quality is a very general term used in power systems, and it has different meaning for different parts of the power system. For the generation side, power quality refers to the protection of generator and generation system. It means to protect the generator from over-rating in terms of real and reactive power. For the transmission side, it means to have high transmission efficiency, less outage, good loadability of transmission lines, requisite flow of power, and balanced sinusoidal quantities over transmission network. For the distribution system, it indicates proper regulation of voltage, compensation of unbalance, harmonics, and undesired variations in voltages and currents. In general, power quality refers to a wide variety of electromagnetic phenomena that characterizes the voltage, current at a given time and location in the power system [1]. There are many variants of power quality definition. As per the Institute of Electrical and Electronics Engineers (IEEE) [2], “Power quality is a concept of powering and grounding sensitive equipment in a matter that is suitable to the operation of that equipment.” The definition predominantly focuses on the issues affecting the operation and performance of the equipment. On the other hand, the International Electrotechnical Commission (IEC 61000-4-30) defines power quality as “Characteristics of the electricity at a given point of an electrical system evaluated against a set of reference technical parameters” [3]. This definition gives more emphasis on measuring and quantifying power system performance [4]–[17]. In a broad sense, power quality refers to maintaining the sinusoidal nature of voltage and current of the fundamental frequency at all points of the power system network. The power quality is also an important aspect of economic consideration, as it saves electrical energy in appliances. In literature, the term power quality is correlated to electromagnetic compatibility (EMC) and electromagnetic interference (EMI) [3]. Also, voltage quality implies current quality and vice-versa, because voltage and current quantities interact through the transmission and distribution line impedance. The product of voltage quality and current quality reflects the quality of the power supply. While the quality of voltage is determined from the source side, the current quality comes from the load side. To illustrate this, consider a balanced three-phase system represented by a single-line diagram in Fig. 1.1.

Figure 1.1 Pertaining to voltage, current, and power quality

In Fig. 1.1, vs is the source voltage which reflects the quality of the source or utility voltage. The current il (t) = is (t) is the load current (also equals to source

Introduction to Power Quality in Power Distribution System

3

current), which reflects the quality of the load current for the given supply voltage, vs . It depends upon the nature of the load, i.e., balance, unbalance and harmonics, etc. Now the voltage vt (t) is the voltage at the load terminal, and the quality of vt (t) depends upon both the source voltage vs (t) and load current il (t). This is clear from the relationship between the vs (t) and vt (t), through the voltage drop across the feeder with resistance Rs and inductance Ls , as given in the following (1.1). vt (t) = vs (t) − Rs il (t) − Ls

dil (t) dt

(1.1)

The instantaneous power pl (t), which is product of voltage, vt (t) and current, il (t), at the load bus is given by, pl (t) = vt (t) il (t)

(1.2)

As clear from (1.1), vt (t) is influenced by the quality of vs (t), il (t), and the value of the feeder resistance Rs and inductance Ls . Also, note that the resistive drop, Rs il (t), and inductive drop Ls didtl (t) depend upon the load current and its rate of change with time, respectively. The latter has a more serious effect on the quality of vt (t) due to the derivative term didtl (t) , if the value of inductance is high. If il (t) contains harmonics, such as the current drawn by a full bridge rectifier or any other converter circuit, it has a worse effect on the quality of vt (t) producing notches, sharp changes, non-sinusoidal variation, and consequently affecting the quality of power at the load terminal, which is a product of voltage, vt (t) and current, il (t), as given in (1.2). Once the terminal voltage is affected, the load current develops more distortion. This goes on till the three-phase load voltages, and load currents settle to certain distorted waveforms, which may not meet the requirement of the connected load resulting in its malfunctioning and erroneous operation. It is observed here that the feeder length and its impedance are important parameters in affecting the power quality at any point between the source and the load. For an ideal source, the connecting feeder to the load should have zero resistance and inductance (Rs = 0 and Ls = 0). Such a system is known as a stiff source. For a non-stiff source, resistance and inductance have some finite non-zero value, leading to a few percentages of voltage drop of the rated voltage. Thus, the extent of “Nonstiffness” of the voltage source is quantified by the magnitude of feeder impedance and its X/R ratio. While non-stiff source has a drawback to affect the power quality at the load bus, at the same time it gives the flexibility to control and regulate the voltage through some compensation schemes at the load bus [18]–[22]. The stiff systems are strong and rigid; therefore, the load has to accept the available voltage quality at its bus. It does not give the flexibility to control the parameters of the affected part of the distribution network. Further, the following points illustrate the nature of the feeder impedance, which helps to correlate the voltage, current, and power quality. 1. The typical value of the per-phase line inductance on a three-phase ac line can be considered to be 1 µH/meter, equivalent to 1 mH/km/phase [5]. 2. Also, it is found that the X/R ratio for the distribution lines is often close to unity, with shunt capacitance largely ignored for calculations [23].

4

Power Quality in Power Distribution Systems: Concepts and Applications

3. From the above observations, we can conclude that the typical value of the distribution feeder impedance can be taken around 0.31+ j0.31 Ω/km/phase. It is to be noted that clear standards have not been established for the length of the distribution line feeders. However, experience suggests that the typical feeder mains of 11 kV level (primary distribution) can run between 1 and 25 km in length. 4. In addition, it is suggested that the length of the low tension secondary distribution lines leading to the customer premises (415 V L-L level) be kept to a maximum of 1 km to avoid considerable voltage drops across the lines. From the above discussion, it is clear that the feeder impedance plays a vital role in co-relating voltage and current and hence power quality problems. For example, if the quality of supply voltage is not good, it will result in poor quality of current on the distribution side, through feeder interaction, as expressed in (1.1). Similarly, poor quality of current on the load side will translate to poor quality of the voltage on the supply side, again through feeder interaction. That is how voltage, current, and power quality relate to each other. Based on this fundamental understanding, the detailed power quality problems and their nature are discussed in the following section.

1.3

POWER QUALITY PROBLEMS

To operate the distribution and transmission network in an acceptable way, the main objective is to maintain power quality in power systems. To achieve this it is important to understand the nature of power quality problems that can occur, how they originate, and their effects on the utility and the consumers. The majority of the power quality problems originate in the distribution grid associated with various kinds of loads used by the end-users. Any deviation in the voltage, current, and supply frequency from its nominal value generates a problem. Thus, it is necessary to define and classify them in a meaningful way. Classification can be done based on the magnitude and duration of the quantities like voltage, current, and frequency. In the following description, power quality issues are classified based on the magnitude and duration of the event. Table 1.1 lists various types of power quality phenomena which are further explained in the following subsections:       

Transients Short-duration or rms variations Long-duration variations Voltage unbalances Waveform distortions Voltage fluctuations Power frequency variations

5

Introduction to Power Quality in Power Distribution System

Table 1.1 Categories and characteristics of power quality phenomena Categories • Transients — Impulsive • Nanosecond • Microsecond • Millisecond — Oscillatory • Low frequency • Medium frequency • High frequency • Short-duration variations — Instantaneous • Sag • Swell — Momentary • Interruption • Sag • Swell • Voltage imbalance — Temporary • Interruption • Sag • Swell • Voltage imbalance • Long-duration variations — Interruption, sustained — Undervoltages — Overvoltages — Current overload • Imbalance — Voltage — Current • Waveform distortion — DC offset — Harmonics — Interharmonics — Notching — Noise • Voltage fluctuations • Power frequency variations

Typical spectral content

Typical duration

5 ns rise 1 us rise 0.1 ms rise

< 50 ns 50 ns–1 ns > 1 ms

< 5 kHz 5–500 kHz 0.5–5 MHz

0.3–50 ms 20 us 5us

0–4 p.u. 0–8 p.u. 0–4 p.u.

0.5–30 cycles 0.5–30 cycles

0.1–0.9 p.u. 1.1–1.8 p.u.

0.5 cycles–3 s 30 cycles–3 s 30 cycles–3 s 30 cycles–3 s

< 0.1 p.u. 0.1–0.9 p.u. 1.1–1.4 p.u. 2%–15%

> 3 s–1 min > 3 s–1 min > 3 s–1 min > 3 s–1 min

< 0.1 p.u. 0.1–0.9 p.u. 1.1–1.2 p.u. 2–5%

> 1 min > 1 min > 1 min > 1 min

0 p.u. 0.8–0.9 p.u. 1.1–1.2 p.u.

steady state steady state

0.5–5% 1–3%

steady state steady state steady state steady state steady state intermittent

0–0.1% 0–20% 0–2%

< 10 s

±0.1 Hz

0–9 kHz 0–9 kHz broadband < 25 Hz

Typical voltage magnitude

0–1% 0.1–7%

6

1.3.1

Power Quality in Power Distribution Systems: Concepts and Applications

TRANSIENTS

Transients are defined as events which are undesired and momentary in nature in the power system. On quantity-specific definition, transient describes a part of the variable that disappears during the transition from one steady state to another. The utility engineers view transients as surges from lightning strokes which are absorbed by surge arrestors to protect the electrical equipments. The end users view transient as anything unusual that might be observed in the power supply including voltage sags, swells, and interruptions. It is characterized by its duration ranging up to about a few tens of milliseconds, and its spectral content, i.e., its rise time and frequency. Transients are further sub-divided into two categories, impulsive and oscillatory, depending upon the waveform of the quantity involved, either current or voltage. 1.3.1.1

Impulsive Transient

Impulsive transients are defined as a sudden, non-power frequency change in the steady-state characteristics of voltage or current or both, where the change or disturbance is unidirectional in nature, either in positive or negative polarity. These transients exhibit similar characteristics to that of a lightning strike. They are normally characterized by their peak value, rise time, and duration of the disturbance.

Figure 1.2 Lightning stroke leading to voltage impulsive transients

For example, an impulsive transient described as 1.2/60 is interpreted in the following manner, 1.2 indicates the rise time of the waveform in microseconds, i.e., the duration in which the magnitude rises from 10% to 90% of its peak value, and 60 indicates the duration in microseconds from the start till the point where the waveform magnitude decays to 50% of its peak value. Depending upon the time range in which a transient rises and its duration, there are three categories of impulsive transients, nanosecond, microsecond, and millisecond. Fig. 1.2 depicts the voltage waveform under the occurrence of a lightning strike along with the zoomed portion of the voltage impulse.

Introduction to Power Quality in Power Distribution System

7

Impulsive transients are typically caused due to lightning strikes and normally damp out quickly due to high-impedance circuit elements. It may sometimes lead to lightning flash-over on power line insulators, thus leading to momentary short circuits. Impulsive transients may give rise to oscillatory transients due to the presence of resonance circuits in the power system. 1.3.1.2

Oscillatory Transient

Oscillatory transients can be defined as a sudden, non-power frequency change in the current and voltage steady-state characteristics, with positive and negative values of the amplitude. In these transients, the electrical quantity involved oscillates multiple times, with its peak magnitude decaying over time. These oscillations are mainly characterized by their magnitude, duration, and spectral content (frequency of oscillations). Based on the frequency of spectral content, transients are further classified as high-, medium-, and low-frequency transients. Table 1.1 mentions the frequency range and duration for each category. Transients with frequency components higher than 500 kHz and a duration measured in few microseconds are considered as high-frequency oscillatory transients. It is caused by some switching events or appears as a local response to an impulsive transient. Transients with frequencies between 5 and 500 kHz and a duration of a few tens of microseconds are considered medium-frequency transients. Energization of back-to-back capacitor banks results in oscillatory current transients [1]. Transients with frequency components less than 5 kHz and duration from 0.3 to 0.5 ms are classified as low-frequency oscillatory transients. These transients occur very frequently in the distribution system due to capacitor bank energization, which results in an oscillatory voltage transient with a frequency range of 300–1600 Hz. This is illustrated in Fig. 1.3. Ferroresonance phenomenon in transformer energization can

Figure 1.3 Low-frequency oscillatory transient caused by capacitor-bank energization

8

Power Quality in Power Distribution Systems: Concepts and Applications

also lead to low-frequency transients [1]. Ferroresonance refers to a situation where the nonlinear magnetic properties of iron in the transformer iron core interact with capacitance in the electrical network to produce a nonlinear tuned circuit with an unexpected resonant frequency. This results in higher values of voltage and current in the transformer. 1.3.2

SHORT-DURATION VARIATIONS

This category deals with power quality problems with anomalies in the voltage magnitude for a duration ranging from 0.5 cycle to 1 min. Short-duration voltage variations are further categorized into three sets based on the duration of the disturbance, instantaneous, momentary, and temporary. Table 1.1 depicts the time duration for all three subcategories. Now looking into the types, there are three types of short duration voltage variations, namely interruption, sag, and swell. 1.3.2.1

Interruption

An interruption occurs when the rms voltage magnitude decreases to less than 0.1 p.u. of its nominal value with a duration ranging from 0.5 cycle to 1 min. They are caused due to faults occurring in the utility or distribution grids. Interruption can occur as a standalone event or can be preceded by a voltage sag. The duration of an interruption depends on the type and capability of the protective device used in the utility system, i.e., faster re-closing of circuit breakers will lead to an instantaneous interruption, and delayed re-closing can lead to momentary or temporary interruptions. Fig. 1.4 depicts a momentary interruption due to a fault, the upper plot shows the instantaneous voltage variation, and the lower plot shows the rms voltage variation.

Figure 1.4 Momentary interruption due to fault and subsequent recloser operation

Introduction to Power Quality in Power Distribution System

1.3.2.2

9

Sag

A voltage sag is characterized as a decrease in the rms voltage magnitude to between 0.1 and 0.9 p.u. of its nominal value, for a duration of 0.5 cycle to 1 min. As per recommended practices, 60% voltage sag refers to the voltage drop to 60% of its nominal value. To avoid any confusion, it is better to specify the nominal voltages as well. Voltage sags are usually caused due to faults in the system, i.e., when there is a fault on a phase, it experiences a sag. It can also be due to the starting of heavy industrial loads like motors, which draws huge inrush currents leading to a voltage dip. Many of us might have experienced this phenomenon in the form of flickering of lights when heavy loads like air-conditioners and water pumps are turned on. Fig. 1.5 depicts a voltage sag caused due to a line-line to ground (LLG) fault. Various sub-classifications of sag depending on the duration are given in Table 1.1.

Figure 1.5 Voltage sag caused by an LLG fault

1.3.2.3

Swell

A volatge swell is characterized by an increase in the rms voltage magnitude above 1.1 p.u., usually ranging from 1.1 to 1.8 p.u., for a duration of 0.5 cycle to 1 min. Like sags, swells can also occur due to faults in the system. When faults occur on a single-phase, there may be a rise in voltage in the other unfaulted phases. It can also result from heavy load rejection in the system or switching on capacitor banks. Fig. 1.6 depicts the voltage swell phenomenon and is shown by instantaneous and rms voltage values. The sub-classification of swell depending on the duration is given in Table 1.1. Swells are very less common events and have varied effects on grounded and ungrounded systems. 1.3.3

LONG-DURATION VARIATIONS

This category of power quality problems consists of deviations in voltage at power frequency for more than 1 min. Long-duration voltage variations are of three types:

10

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 1.6 Voltage swell

overvoltage, undervoltage, and sustained interruptions. This category is similar to short-duration voltage variations; however, they are caused due to changes in the system load but not as a result of system faults. 1.3.3.1

Overvoltages

1.3.3.2

Undervoltages

1.3.3.3

Sustained Interruptions

Overvoltages are kind of voltage swells with longer duration. They are characterized by an increase in the rms voltage magnitude above 1.1 p.u., usually ranging from 1.1 to 1.2 p.u., for duration more than 1 min. They are caused due to switching-off of large loads or as a result of a change in the reactive power compensation on the system, i.e., injection of more reactive power into the system or due to poor voltage regulation controls.

Undervoltage is similar to voltage sag with a longer duration. They are characterized by a decrease in the rms voltage magnitude to less than 0.9 p.u., usually ranging from 0.8 to 0.9 p.u., for duration more than 1 min. They are caused by the opposite events that cause overvoltages, like switching-on of heavy loads or capacitor banks, or a change in reactive power compensation of the system, i.e., absorption of reactive power from the system.

This type of long-duration variation is similar to interruptions. These are characterized by the decrease in rms voltage magnitude to less than 10% of its nominal value, for a duration longer than 1 min. They are mainly caused by permanent faults in the system, which can be cleared by manual intervention.

Introduction to Power Quality in Power Distribution System

1.3.4

11

IMBALANCE

This category of power quality problem deals with the anomaly in the three-phase sequence components of an ac quantity. As per the IEEE definitions, imbalance in a three-phase system is defined as the ratio of negative to positive sequence component magnitudes, given as a percentage. Depending upon the quantity involved, there can be current or voltage imbalances. Generally, the number of current imbalances are much higher than those of voltage imbalances. Voltage imbalances of 2% or less are mainly caused due to unbalanced single-phase loads on a three-phase circuit. Severe voltage unbalances, those greater than 5%, result from single phasing conditions such as disconnection of one phase of a three-phase motor. Current imbalances are also a result of the presence of single-phase loads. 1.3.5

WAVEFORM DISTORTION

The problems discussed till now deal with deviations in the ac quantity’s magnitudes involved, either voltage or current. However, for an ac quantity to be perfectly ideal, its waveform has to be sinusoidal in nature. Therefore, this category of power quality problems deals with the types of anomalies in the waveform of the quantity involved. There are mainly five types of waveform distortion, dc offset, harmonics, interharmonics, notching, and noise. 1.3.5.1

DC Offset

1.3.5.2

Harmonics

The presence of a dc voltage or current in any ac voltage or current is defined as a dc offset. These problems can result from geomagnetic disturbance on transmission lines or asymmetric operation in power electronic converters, which leads to the undue effect of half-wave rectification. The presence of dc in the ac power system can lead to transformer saturation and associated heating, causing stress in the insulation material.

When ac quantities like voltage and current have frequencies that are integer multiples of the system’s fundamental frequency, they are termed harmonics. Harmonics, when combined with the fundamental frequency component, leads to waveform distortion. Power electronics converters involving continuous switching are the major source of harmonics. Devices like diode bridges and line-commutated converters can be seen as harmonic current sources. Other devices like pulse width modulated inverters can be seen as harmonic currents reflected as harmonic voltages. As more and more power electronics switching devices are being incorporated into the power system, harmonics are also increasing, which has been a growing concern for both the customers and the utility. Harmonics are characterized by a parameter known as the Total Harmonic Distortion (THD), which is defined as the ratio rms magnitude of the harmonic component to the fundamental component. In many drive applications under lightly loaded conditions, THD gives erroneous results for current harmonics.

12

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 1.7 Current waveform and harmonic spectrum for a power electronic-based load

IEEE has defined another parameter for current harmonics known as Total Demand Distortion (TDD) to avoid such cases. Fig. 1.7 depicts the waveform and harmonics spectrum for phase-a input current of three-phase full bridge rectifier circuit. 1.3.5.3

Interharmonics

1.3.5.4

Notching

In similar notions of harmonics, if the ac quantity has frequency components other than fundamental and is not an integer multiple of the fundamental frequency, they are called as interharmonics. They are mainly caused by devices whose control is not synchronized with the fundamental power frequency like pulse-modulated inverters, cycloconverters, induction furnaces, and arcing devices.

Notching is a kind of voltage disturbance resulting from events occurring in the current waveform, which leads to large periodic spikes in the voltage waveform. It is caused by the normal operation of power electronic converters when switching operation leads to current commutation from one phase to the other, causing a momentary short circuit between the two phases. Notching is periodic and has high-frequency content, which makes a part of transients as well. Voltage notching can cause frequency or timing errors in phase-locked-loop systems due to the detection of additional zero crossings. Fig. 1.8 illustrates an example of voltage notching due to converter operation.

Introduction to Power Quality in Power Distribution System

13

Figure 1.8 Example of voltage notching by converter operation

1.3.5.5

Noise

Distortion or disturbance of any kind that does not fit into transients or any other category of waveform distortion can be categorized as noise. These are unwanted electrical signals with spectral content usually less than 200 kHz and magnitude usually less than 1% of the signal magnitude. These are superimposed on voltage or current signals in phase conductors and are passed on to the neutral conductors. These can be caused by power electronic devices, arcing devices, switching power supplies, control circuits, and improper grounding. The use of filters, isolation transformers, and line conditioners can minimize the effect of noise. 1.3.6

VOLTAGE FLUCTUATIONS

We have observed that for sag or undervoltage, the rms magnitude goes below 0.9 p.u., or in swell or overvoltage, the rms magnitude goes above 1.1 p.u. Now if the voltage magnitude variations are below 1.1 p.u. and above 0.9 p.u., they fall in this category of voltage fluctuations. As per the IEEE definition, voltage fluctuations are defined as variations in the voltage envelope or a series of random changes in the magnitude, which does not exceed the limits of 0.95-1.05 p.u. Voltage fluctuations can also be seen as a form of illumination change in lamps. Voltage fluctuations can be caused by any cyclic variations in the loads’ reactive power component or continuous variations in the load current magnitude, such as arc furnaces, welding machines, etc. A decrease in voltage can also be a result of poor load power factor, leading to high reactive current and subsequently more feeder drop. 1.3.7

POWER FREQUENCY VARIATIONS

These are deviations in the fundamental frequency from its predefined nominal value, like 50 or 60 Hz. The nominal value of power frequency is decided solely by the constructional features of the generators producing power. As long as the balance of power generation versus load demand is maintained, the frequency is stable. The only cause of power frequency variation is the shift in the power balance. The magnitude

14

Power Quality in Power Distribution Systems: Concepts and Applications

and duration of frequency variations depend upon the dynamic load characteristics and generation system. There is always a minor change in frequency going on, which is acceptable, however severe changes are very rare but highly fatal to the power system as they can lead to complete grid blackouts.

1.4

MITIGATION TECHNIQUES

In the previous section, we discussed various power quality problems. For the efficient and reliable operation of the power system, it is desirable to eliminate or minimize them. Earlier, power quality problems were classified into various categories and subcategories. We now realize that almost all power quality problems arise due to non-sinusoidal variations of voltage, or current over a certain period of time. Thus, it is very likely that every voltage or current quantity should retain the fundamental sinusoidal component after compensation. To achieve this, there are two possible approaches. One approach is to eliminate the problem at the source, i.e., eliminate the cause, leading to such unwanted components. This is achieved through network re-configuration devices. The second approach is to employ different mitigating devices to take care of the unwanted features by injecting desired voltage and currents. These mitigating devices are referred to as Custom Power devices. Based on their functionality, these are broadly divided into two types: network re-configuring type and compensating type. Applications related to various mitigation techniques are listed in Table 1.2. 1.4.1

NETWORK RE-CONFIGURING TYPE

Network re-configuring type devices nowadays used are based on solid-state switches like GTOs or thyristors, unlike the previously used mechanical switches, which were replaced due to their slow operations. They are normally used for fast current limiting and current interruption during faults. There are mainly three devices under this category [9]. 1. Solid-State Current Limiter (SSCL): Under a fault condition, this Gate Turn Off Thyristor (GTO) based device inserts a fault current limiting inductance in the circuit, as illustrated in Fig. 1.9. The zinc oxide arrestor (ZnO), current limiting inductor, and GTO units are connected in parallel. The combined circuit is inserted in series with the line. When fault takes place, the current rises to a higher magnitude. This is sensed by the current sensing unit, and accordingly, the turn-off command is sent to the GTOs by applying a negative gate pulse. This will reduce the current through the main line in a matter of few microseconds. During switching-off of the GTOs, the snubber circuit across each GTO switch ramps up the voltage to the level clamped by the ZnO arrestor. Subsequently, it allows the linear rise of current through the inductor till it reaches its voltage limit. Once the fault clears, the turn-on command is given to the GTO units. Thereafter, the current resumes to its rated value. The inductor and ZnO arrestor become non-functional. Several GTOs can be connected back to back in a series to support bidirectional current flow and to match voltage rating.

Introduction to Power Quality in Power Distribution System

15

Table 1.2 Different types of custom power devices and their applications Custom power devices Applications Solid-state current limiter/ Solid-state circuit breaker Solid-state transfer switch Distribution static compensator (DSTATCOM) Dynamic voltage restorer (DVR)

Unified power quality conditioner (UPQC)

Fault current limitation Fault current interruption with further reclosing Protection against voltage sag and swell Transfer of load from one feeder to another Load current balancing Power factor correction Current harmonic compensation Voltage sag and swell protection Load bus voltage regulation Voltage balancing Voltage Flicker attenuation VAr compensation Voltage regulation Voltage and current balancing Voltage and current harmonic suppression Active and reactive power control

2. Solid-State Circuit Breaker (SSCB): Circuit Breakers are generally mechanical in nature, but SSCB uses a combination of GTOs and Thyristors with the capability to interrupt fault currents very rapidly and perform an auto-reclosing function, as shown in Fig. 1.10. The device allows the fast transfer of power from one supply source to another. The circuit configuration of SSCB is similar to SSCL, except that it uses additional anti-parallel connected thyristor switches in series with the inductor, which eventually allows the cut-off the

Figure 1.9 Solid-state GTO based current limiter

16

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 1.10 Solid-state circuit breaker

inductor current developed through inductor during switching off of GTOs. The number of switches used in the circuit breaker are determined based on its voltage and current ratings. 3. Solid-State Transfer Switch (SSTS): Solid-state transfer switches are thyristorbased devices used for protecting sensitive loads from voltage sag, swell, and interruption. They can rapidly transfer the load from the main source to an auxiliary source within few milliseconds (half a cycle) whenever a voltage sag, swell, or interruption occurs in the primary or main line, acting as an uninterrupted power supply. The schematic diagram of the SSTS is shown in Fig. 1.11. There are two switches: Switch 1 is on the main feeder side, and Switch 2 is on the auxiliary feeder side. Each switch comprises of a pair of series connected thyristors (to match the voltage rating) in an anti-parallel connection. A ZnO arrestor is connected across each switch to provide the voltage limit. Under normal condition the main feeder supplies the load. When fault takes place on the main feeder side, it results in voltage sag, swell, or

Figure 1.11 Solid-state transfer switch

Introduction to Power Quality in Power Distribution System

17

interruption, and the load is immediately connected to the auxiliary feeder in a make before break action. The currents imp , imn and iap , ian are positive, negative currents on main and auxiliary side of the feeders, respectively. 1.4.2

COMPENSATING TYPE

There are broadly two categories of compensating type devices, namely passive and active compensating devices. Passive compensation is used to mitigate power quality problems by using lossless passive components such as capacitors and inductors. A schematic of a passive filter is shown in Fig. 1.12. In the figure, the current i f refers to the compensator or filter injected currents at the point of common coupling (PCC) to alleviate unwanted components of load current, il , resulting in desired source current, is . Inductance-resistance, Ls − Rs represents the feeder impedance, Zs . Passive compensation methods can achieve the following objectives: reactive power compensation, voltage regulation, and power factor correction. However, these are not suitable for higher order harmonics and dynamic compensation.

Figure 1.12 Passive filter

Active compensating devices also called as active power filters, which are used for active filtering, load balancing, power factor correction, and voltage regulation with fast dynamic characteristics. There are mainly three types of active compensating devices which we will learn in the subsequent chapters. 1. Distribution Static Compensator (DSTATCOM): This is a shunt type compensating device, also called as shunt active power filter. It injects current at the PCC such that the compensated current nullifies the unbalance and harmonics in the load current. The scheme is mainly used for power factor correction, load balancing, and harmonic eliminations. As a result of this, it improves the quality of the bus voltage. A schematic of DSTATCOM is shown in Fig. 1.13. It consists of a voltage source inverter (VSI) supported by dc link voltage. The VSI is connected to the PCC through interface inductor, L f . 2. Dynamic Voltage Restorer (DVR): This is a series type compensating device that injects a voltage such that the terminal voltage at the load remains free of voltage fluctuations, sag, swell, interruptions, and harmonic distortions. Thus,

18

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 1.13 Distribution static compensator (DSTATCOM)

Figure 1.14 Dynamic voltage restorer (DVR)

its main objective is to regulate the voltage across the load terminal bus. The schematic of DVR is shown in Fig. 1.14. It consists of a voltage source inverter supported by the dc link voltage. The output terminals of the inverter are connected to the primary of the transformer and the secondary of the transformer injects voltage in series with the line. The transformer’s inductance, along with the filter components shapes the injected voltages, v f , following the reference DVR voltages. 3. Unified Power Quality Conditioner (UPQC): This is a combination of DSTATCOM as shunt active power filter and DVR as series active power filter. Thus, it combines the advantage of both the shunt and series compensation. A schematic of UPQC is shown in Fig. 1.15. The device UPQC regulates the voltage at the load bus by injecting series voltage and maintains the supply current by injecting current at the PCC. It has two voltage source inverters and a common dc link to support the operation of the UPQC. Since the dc link is

Introduction to Power Quality in Power Distribution System

19

Figure 1.15 Unified power quality conditioner (UPQC)

common, its proper design and rating are very important for satisfactory voltage and current compensation. Its control is more complicated than the other custom power devices as it involves two inverters along with their coordination with the dynamics of the dc link voltage.

1.5

POWER QUALITY MONITORING

Power quality problems are randomly occurring events, and if timely detection and mitigation are not performed can lead to the malfunctioning of critical and sensitive load, which can further cause substantial financial losses. Therefore, monitoring power quality events is an essential part of the mitigation process. There are many aspects of monitoring a power quality event. Identifying events specific to each location of the power system (i.e., a disturbance can have different effects, at a given location as the distance from the point of occurrence changes). Further, power quality monitoring allows to evaluate and predict the performance of loads and choose appropriate mitigation techniques. Table 1.3 depicts some of the parameters used for the detection of power quality events [24]. In general, a power quality monitoring system comprises of four stages, 1. Data Acquisition: In data acquisition, the measured line voltages and currents are sampled and converted to their equivalent digital signals using analog to digital converters (ADCs). 2. Characterization: In characterization, the digital signals of current and voltage are used to determine different parameters using various signal processing algorithms, and are stored. 3. PQ Analysis: At analysis stage, the calculated parameters are analyzed and compared against defined standards using some processing units to detect power quality events and disturbances. 4. Statistical Analysis: In this stage, statistical analysis of the detected power quality events along with previous trends of those events and parameters is performed, and appropriate mitigating actions are selected.

20

Power Quality in Power Distribution Systems: Concepts and Applications

Table 1.3 Stored data used for monitoring various PQ events Power quality events Characteristics Transients

Short-duration voltage variation (Sag, swell, and interruption) Long-duration voltage variation (Undervoltage, overvoltage) Unbalance Waveform distortions

Voltage fluctuations

Power frequency variations

1.6

Rate of rise, peak amplitude, duration, spectral content, energy potential, mean absolute variation squared amplitude (MAVSA) index Duration, rms magnitude (minimum / or maximum), and clearing time Duration, rms magnitude (minimum or maximum), and clearing time Duration, percentage unbalance Harmonic spectrum with magnitude and phase angles, total harmonic distortion (THD), total demand distortion (TDD), frequency spectrum of notching, minimum / maximum amplitude of notch Duration, rms magnitude, instantaneous flicker (IFL), perceptibility short term (Pst), perceptibility long term (Plt) Duration, frequency (minimum / maximum)

POWER QUALITY INDICES

Various power quality indices are used to quantify and characterize the disturbance in electrical systems, industrial plants or residential loads. These indices serve to indicate the complex time and frequency domain waveform into a single number. The important indices used to assess the power quality are listed in Table 1.4 [25]. The indices mentioned in the table are calculated using standardized procedures and are meaningfully interpreted by practicing engineers. The important power quality indices are described in this section for a periodic voltage or current waveform, x(t), given as following. x(t)

=

√ ∞ √ Xdc + 2X1 sin(ωt − φ1 ) + 2 ∑ Xh sin(hωt − φh ) ∞

= Xdc + x1 + ∑ xh

h=2

(1.3)

h=2

In the above equation, x(t) represents the instantaneous voltage or current quantity. The terms, Xdc , X1 , and Xh are dc component, rms value of fundamental, and

Introduction to Power Quality in Power Distribution System

Table 1.4 Commonly used power quality indices Index Definition  Total harmonic 2 ∑∞ i=2 Ii /I1 distortion (THD)  Total demand 2 ∑∞ i=2 Ii /IL distortion (TDD) Power factor (PF) Ptot /(|Vrms | |Irms |)  Telephone 2 ∑∞ i=2 wi Ii /Irms influence factor  2 2 IT product ∑∞ i=2 wi Ii  2 2 VT product ∑∞ i=2 wi Vi  ∞ 2 2  ∞ 2 K factor ∑h=1 h Ih ∑h=1 Ih Crest factor Vpeak /Vrms Unbalance factor V− /V+ Flicker factor

∆V /|V |

21

Application(s) Measure of waveform distortion Measure of waveform distortion Energy metering Audio circuit interference Audio circuit interference, Shunt capacitor stress Voltage distortion index Transformer derating Insulation capacity Degree of three-phase unbalance Incandescent lamp operation, voltage regulation, short circuit capacity

harmonics, respectively. The terms h, ω, t, and φh represent the harmonic number, angular frequency, time variation, and phase angle of the sinusoidal voltage or current waveform. The dc component, rms, and average values of sinusoidal waveforms are denoted by capital letters such as Xdc , Xrms , Xavg (terms explained below). The instantaneous ac quantities are denoted by small letters such as x1 , x2 , ....xh as indicated in (1.3). Harmonics: The sinusoidal voltage or current waveform with frequency being integer multiple of fundamental frequency. Odd integer and even integer multiples result in odd and even harmonics. In (1.3), x1 , x2 , ...xh components represent the fundamental and harmonic components. Interharmonics: Sinusoidal voltage or current waveform with frequency which are not integer multiple of fundamental frequency. Subharmonics: Sinusoidal voltage or current waveform with frequency below the fundamental frequency.

22

Power Quality in Power Distribution Systems: Concepts and Applications

Average value of voltage or current waveform: The average value of a periodic nonsinusoidal voltage or current waveform is defined as following. Xavg =

1 T

 T 0

x(t) dt

(1.4)

where x(t) is instantaneous voltage or current quantity and T is the time period of fundamental component of x(t). rms value of voltage or current waveform: The rms value of a periodic nonsinusoidal voltage or current waveform is defined as following.   1 T 2 Xrms = x (t) dt (1.5) T 0 Crest Factor (CF): It is a measure of ratio of the peak value of voltage or current quantity to its rms value. It is defined as follows. Crest Factor (CF) =

X peak Xrms

(1.6)

Form Factor (FF): Form factor indicates the shape of the voltage or current waveform. It is defined as the ratio of the rms to the average value. Form Factor (FF) =

Xrms Xavg

(1.7)

Ripple Factor (RF): Ripple factor is a measure of the ripple content of the voltage or current waveform. It is defined as the ratio of rms value of ac component to the dc component of voltage or current quantity. Xac (1.8) Xdc  2 − X 2 which leads to RF = = Xrms dc

Ripple Factor (RF) = 2 = X 2 + X 2 , it implies that X Since Xrms ac ac dc √ 2 FF − 1.

Harmonic Factor (HF): Harmonic factor of hth harmonic is the ratio of the rms values of the hth harmonic to the fundamental. Thus, Harmonic Factor (HF) =

Xh X1

(1.9)

Introduction to Power Quality in Power Distribution System

23

Total Harmonic Distortion (THD): Total Harmonic Distortion (THD) is the most commonly used index to quantify the harmonic content of the non-sinusoidal waveform. It is defined as the ratio of the rms value of all the harmonic components (XH to the rms value of the fundamental (X1 ). It is given below. ∞ ∑ Xh2 XH h=2 = (1.10) Total Harmonic Distortion (THD) = X X1 1   X 2 − X12 X 2 = −1 = X1 X1 Total Demand Distortion (TDD): Total Demand Distortion (TDD) is similar to THD except that it is expressed as the ratio of the rms value of the total harmonic components (XH ) to the rms value of the waveform, x(t). It is given as follows. Total Demand Distortion (TDD) =

Xh X

(1.11)

Power Factor (p f ): Power Factor (p f ) is defined as the ratio of real power (P) to the apparent power (S). For a given voltage, v(t) and current, i(t), as given in (1.3), with x(t) representing voltage, v(t) or current, i(t). The power factor is defined as follows. pf =

P = S

1 T T 0

v(t) i(t) dt Vrms Irms

(1.12)

Telephone Influence Factor (TIF): Telephone Influence Factor (TIF) determines the influence of the power system harmonics on telephone communication systems. It is the ratio of weighted rms to the rms value of the quantity and is expressed as below.  2 ∑∞ i=2 wi Xi (1.13) T IF = Xrms C-Message Weights: C-Message is similar to TIF, except that the weights wi are replaced with weights ci . In C-message, C stands for computer. This index is also called as Computer Interference Factor (CIF). It determines the impact of power system harmonics on telephone communication systems. It is the ratio of weighted rms to the rms value of the quantity and is expressed as below.  2 ∑∞ h=2 wi Xi T IF = (1.14) Xrms

24

Power Quality in Power Distribution Systems: Concepts and Applications

IT and VT Product (IT, VT): IT (current product) and VT (voltage product) are weighted rms values of harmonic components of current and voltage, respectively. It is basically the product of rms value of current or voltage with its TIF (hence the terms, I ∗ T or V ∗ T ). These are kind of alternative indices to the THD. The IT and VT are defined as below.  IT

VT



∑ w2i Ii

=

=



(1.15)

i=2 ∞

∑ w2i Vi

(1.16)

i=2

K-Factor: K-factor is the ratio of sum of weighted squared values of the harmonic load currents according to their effects on transformer heating to the sum of squared rms components. The higher K-factor, implies more heating effect due to harmonic components. Some other loads such as computers, solid-state devices, and motors which generate non linear load currents, cause transformers and system neutrals to overheat, reducing their life span and eventual failure of the transformer. The Kfactor is defined as below. h=h

K- f actor =

∑h=1max h2 Ih2 h=h

∑h=1max Ih2

(1.17)

Unbalance Factor: Unbalance factor is used to measure the unbalance present in the three-phase voltages or currents. This is expressed as the ratio of the negative sequence, (X − ) to positive sequence, (X + ) components as given below. Unbalance Factor =

X− X+

(1.18)

Flicker Factor: It is defined as the ratio of change in voltage to its rms value, as expressed below. Flicker Factor =

1.7

∆V V

(1.19)

POWER QUALITY STANDARDS

In the preceding sections, we discussed power quality problems and their mitigation devices. However, due to practical constraints, it is not always possible to completely eliminate power quality problems. Thus, it is imperative to alleviate the problems and maintain certain permissible limits of various parameters. If the parameters are beyond the safe operating limits, these start affecting consumers, utilities, and therefore it can be a matter of concern. To address this issue, many regulatory bodies

Introduction to Power Quality in Power Distribution System

25

such as the Institute of Electrical and Electronics Engineers (IEEE), International Electrotechnical Commission (IEC), American National Standards Institute (ANSI), British Standards (BS), and Computer Business Equipment Manufacturers Association (CBEMA), etc., have developed certain sets of power quality standards which are used by the electric utilities in generation, transmission and distribution systems to maintain power quality within acceptable levels. The important limits for voltage and currents at different voltage levels are given in [5]. The allowable voltage limits in the power system are normally within ±10% over the rated value. The allowable harmonics in voltage measured as THD should be below 8%, and for the current, THD should be below 5%. When short circuit capacity is high, slightly more distortions can be allowed from these THD values for the voltage and current. The dc offset of the load currents should be avoided as it leads to the saturation of the transformer and electrical machines leading to their inefficient operation. The dc offset also leads to the misoperation of electronic switching by affecting zero crossing and other control logic. Similarly, there are standards for allowable frequency variations. As per IEEE 1547-2018, allowed frequency variation is normally within ±0.02 p.u. Some of the important power quality standards are mentioned below. IEEE Power Quality Standards • IEEE 1159: Monitoring Electric Power Quality • IEEE 1159.1: Guide For Recorder and Data Acquisition Requirements • IEEE 1159.2: Power Quality Event Characterization • IEEE 1159.3: Data File Format for Power Quality Data Interchange • IEEE 1564: Voltage Sag Indices • IEEE 1346: Power System Compatibility with Process Equipment • IEEE 1100: Power and Grounding Electronic Equipment (Emerald Book) • IEEE 1433: Power Quality Definitions • IEEE 1453: Voltage flicker • IEEE 519: Harmonic Control in Electrical Power Systems • IEEE 519A: Guide for Applying Harmonic Limits on Power Systems • IEEE 446: Emergency and standby power • IEEE 1409: Distribution Custom Power • IEEE 1547: Distributed Resources and Electric Power Systems Interconnection

26

Power Quality in Power Distribution Systems: Concepts and Applications

IEC Power Quality Standards • IEC Standard 61000-4-7: Deals with the requirements for monitoring and measuring harmonics. • IEC Standard 61000-4-15: Describes the instrumentation and procedures for monitoring flicker. (The IEEE flicker task force working on Standard IEEE P1453 is set to adopt the IEC Standard 61000-4-15 as its own.) • IEC Standard 61000-4-30: Plans on providing overall recommendations for monitoring all types of power quality phenomena. • IEC Standard 61000-2-8: Environment—Voltage Dips and Short Interruptions. • IEC Standard 60034-1: Rotating electrical machines, Part 1: Rating and performance

1.8

CBEMA AND ITIC CURVES

Computer and Business Equipment Manufacturers’ Association, 1970 (CBEMA), Information Technology Industry Council (ITIC) curves are commonly used to indicate power quality in relation to the operation of the electrical equipments [26]. These curves are also known as voltage acceptability or tolerance curves, as shown in Fig. 1.16. The curves basically indicate the tolerance of mainframe computer equipment to the magnitude and duration of the applied voltage. The x-y axes represent the time duration and magnitude of the applied voltage for the equipment, respectively. The key observation from these curves is that under disturbance conditions, there is an energy impact, which here is represented as product of square of rms voltage, V 2 and the time duration, ∆t. Thus, every equipment has a certain maximum energy impact that it can tolerate under disturbances. If the impact of the energy level is more,

Figure 1.16 CBEMA and ITIC curves

Introduction to Power Quality in Power Distribution System

27

then the operation of the load is disrupted, which is indicated by the outer regions of the curves. The inner regions of the curves indicate the acceptable power range. It can further be observed that 100% change can be normally accepted for about a millisecond, and ±10% is acceptable for the continuous operation of the equipment.

1.9

SUMMARY

In this chapter, the concept of power quality in power systems has been discussed with a focus on distribution power systems. The relation of quality of voltage, current, and power is explained using simple expressions and circuitry. Various power quality problems are discussed and presented with their classification, causes, and effects on the end-users. Further, to eliminate these power quality disturbances, several mitigation techniques like network reconfiguring and compensating typle devices have been discussed, and their operating principles are explained. For monitoring of power quality aspects using power quality analyzers, important power quality indices have been defined and explained. Power quality problems are an inherent part of any electrical network which arises due to the use of certain kinds of electrical equipment and situations within the network. Therefore, it is not possible to completely eliminate all the problems rather maintain the electrical quantities within prescribed limits for which various power quality standards developed by international organizations like IEEE, IEC, ANSI, BS, and CBEMA are described. The standards providing the guidelines to allowable limits of magnitude and harmonic components of voltage and system frequency variations are discussed and explained. Further, CBEMA and Information Technology Industry Council (ITIC) curves illustrate the safe operating range for electrical products and systems.

1.10

PROBLEMS

P 1.1 What is power quality? What does the term power quality refer to in a power distribution system? P 1.2 Explain how the feeder impedance affects power quality at the load bus. P 1.3 What are stiff and non-stiff sources? Relate them to the X and R values of the feeder impedance. P 1.4 Explain how voltage related power quality becomes a current related problem and vice-versa. How the quality of power is related to the quality of voltage and current? P 1.5 List various power quality problems in the power distribution network. Describe each one of them in detail. P 1.6 What is the basis of the classification of short term and long term variations of voltage? P 1.7 Define voltage sag, swell, and interruption along with their classification on time duration.

28

Power Quality in Power Distribution Systems: Concepts and Applications

P 1.8 Describe harmonics and interharmonics in the power system. List down their causes and effects. P 1.9 What is voltage fluctuation? List a few loads causing voltage fluctuation and explain how these loads generate voltage fluctuation in the power system. P 1.10 Why is power quality monitoring important for industrial plants/loads? What are the various stages involved in power quality monitoring? P 1.11 Describe power quality indices and list the important ones to characterize the power system’s performance. P 1.12 Define the following terms: (a) harmonics, (b) interharmonics, (c) subharmonics, (d) harmonic factor, (e) total harmonic distortion (THD), (f) total demand distortion (TDD), (g) power factor, and (h) unbalance factor. P 1.13 What is the significance of power quality standards? List the important standards concerning power quality. P 1.14 Discuss the CBEMA and ITIC curves and explain their significance. From these curves, what do you infer about the magnitude and duration of the voltage supplying an electrical device?

REFERENCES 1. IEEE Standard 1159–2019, “IEEE recommended practice for monitoring electric power quality,” 2019. 2. IEEE Standard 1100–2005 (Revision of IEEE Std 1100–1999), “IEEE recommended practice for powering and grounding electronic equipment,” 2006. 3. IEC, “IEC 61000-3-2 (1995) Electromagnetic Compatibility (EMC) Part 3: Limits- Section 2 Limits for Harmonics Current Emissions (Equipments input currents ¡ 16 A per phase),” 1995. 4. IEEE Standard 1459–2010 (Revision of IEEE Std 1459–2000), “IEEE standard definitions for the measurement of electric power quantities under sinusoidal, nonsinusoidal, balanced, or unbalanced conditions,” 2010. 5. IEEE Standard 519–2014 (Revision of IEEE Std 519–1992), “IEEE recommended practices and requirements for harmonic control in electrical power systems,” 2014. 6. W. Shepherd and P. Zand, Energy Flow and Power Factor in Nonsinusoidal Circuits. Cambridge University Press, 1979. 7. M. H. J. Bollen, Understanding Power Quality Problems: Voltage Sags and Interruptions. Wiley-IEEE Press, 1999.

Introduction to Power Quality in Power Distribution System

29

8. Math H. J. Bollen and Irene Y. H. Gu, Signal Processing of Power Quality Disturbances. Kluwer Academic Publishers, 2006. 9. Arindam Ghosh and Gerald Ledwich, Power quality enhancement using custom power systems. Wiley, 2006. 10. Bhim Singh, Ambrish Chandra, and Kamal Al-Haddad, Power Quality Problems and Mitigation Techniques. Wiley, 2015. 11. Roger C. Dugan, Mark F. McGranaghan, Surya Santoso, and H. Wayne Beaty, Electrical Power System Quality. Tata McGraw–Hill, 2008. 12. M. A. Masoum and E. F. Fuchs, Chapter 1 – Introduction to Power Quality, M. A. Masoum and E. F. Fuchs, Eds. Academic Press, 2015. 13. J. Schalabbach, D. Blume, and T. Stephanblome, Voltage Quality in Electrical Power Systems. The Institute of Engineers, London, UK, 2001. 14. S. Chattopadhyay, M. Mitra, and S. Sengupta, Electric Power Quality. Springer, 2011. 15. A. Baggini, Ed., Handbook of Power Quality. Wiley Sons, England, 2008. 16. B. W. Kennedy, Power Quality Primer. Springer, 2000. 17. C. Sankaran, Power Quality. CRC Press, 2002. 18. Mahesh K. Mishra, A. Ghosh, and A. Joshi, “Load compensation for systems with non-stiff source using state feedback,” Electric Power Systems Research, vol. 67, no. 1, pp. 35–44, 2003. 19. Mahesh K. Mishra, A. Ghosh, and A. Joshi, “Operation of a DSTATCOM in voltage control mode,” IEEE Transactions on Power Delivery, vol. 18, no. 1, pp. 258–264, 2003. 20. L. P. Kunjumuhammed and Mahesh K. Mishra, “A control algorithm for singlephase active power filter under non-stiff voltage source,” IEEE Transactions on Power Electronics, vol. 21, no. 3, pp. 822–825, 2006. 21. C. Kumar and Mahesh K. Mishra, “A multifunctional DSTATCOM operating under stiff source,” IEEE Transactions on Industrial Electronics, vol. 61, no. 7, pp. 3131–3136, 2014. 22. C. Kumar, Mahesh K. Mishra, and M. Liserre, “Design of external inductor for improving performance of voltage-controlled DSTATCOM,” IEEE Transactions on Industrial Electronics, vol. 63, no. 8, pp. 4674–4682, 2016. 23. Brendan Fox, L. Bryans, D. Flynn, N. Jenkins, D. Milborrow, M. O’Malley, R. Watson, and O. Anaya-Lara, Wind Power Integration: Connection and System Operational Aspects (2nd Edition). IET Power and Energy Series, 2014.

30

Power Quality in Power Distribution Systems: Concepts and Applications

24. Dong-Jun Won, Il-Yop Chung, Joong-Moon Kim, Seung-Il Moon, Jang-Cheol Seo, and Jong-Woong Choe, “Development of power quality monitoring system with central processing scheme,” vol. 2, pp. 915–919, 2002. 25. G. T. Heydt, Electric Power Quality. Stars in a Circle Publications, 1991. 26. Information Technology Industry Council (ITIC), 1250, Eye Street NW, Suite 200 Washington, D.C. (http://www.itic.org).

Circuits: 2 Single-Phase Power Definitions and Components

2.1

INTRODUCTION

Single-phase circuits are frequently used in low power applications such as lighting, heating, single-phase motors and few large electric motors in residential and small or medium size industries. The single-phase circuits can be used as part of three-phase circuits or as an independent single-phase system. The single-phase load may form about 10–20% of the total load which is a significant portion to affect the overall performance of the power system. Most single-phase applications assume voltage and current waveform as sinusoidal. When these waveforms are not sinusoidal, the errors may occur in their measurements and in the operation of equipments. Under these conditions, it is necessary to evaluate the measuring techniques used in power system with the increased power electronic-based loads and other non-linear loads. Therefore, the definitions of power and its various components are very important to understand the quantitative and qualitative power quality aspects in power system [1]–[18]. Some quantities, such as reactive power, power factor under non-sinusoidal voltage and current, are not properly elaborated and explained in the literature. The importance of proper definition and quantification of various power terms and components becomes more meaningful in load compensation, voltage regulation, as well as measurement of electrical energy using digital energy meters, when there are harmonics in voltage and current. In this chapter, an attempt is made to explain the concept of instantaneous power, instantaneous active power, instantaneous reactive power, real power, reactive power, and power factor in a systematic and connecting manner. This is not only necessary from the point of view of conceptual clarity, but also very much required for practical applications such as metering, quantification of active power, reactive power, power factor, and other power quality parameters in power system. These aspects become more important when power system is not ideal, i.e., when system deals with unbalance, harmonics, faults and fluctuations in voltage and frequency. We therefore, in this chapter, explore the concepts and fundamentals of single-phase system with some practical applications and illustrations.

2.2

POWER TERMS AND DEFINITIONS IN SINGLE-PHASE SYSTEMS

Let us consider a single-phase system with sinusoidal voltage supplying a linear load as shown in Fig. 2.1(a). A linear load is one which consists of ideal resistive, inductive and capacitive elements. The instantaneous voltage and current are expressed as DOI: 10.1201/9781032617305-2

31

32

Power Quality in Power Distribution Systems: Concepts and Applications

below. v(t) = i(t) =





2V sin ωt 2 I sin(ωt − φ )

(2.1)

The corresponding voltage √ √ and current waveforms are shown in Fig. 2.1(b) with peak values Vm = 2V, Im = 2 I and frequency, ω = 2 π f for f = 50 Hz with V and I are the rms value of voltage and current, respectively. The current lags voltage by φ degrees, which depends on the load parameters, i.e., resistance, inductance, and capacitance values. The voltage in the above equation is related to the work done per unit of charge, and the current is related to the rate of change of charge, i.e., v = dW /dq, and i = dq/dt. Therefore, the product of the voltage and current is the rate of work done, i.e., v i = dW /dt J/s or Watts. The instantaneous power can be expressed as following.

Figure 2.1 (a) A single-phase system and (b) voltage and current waveforms

p(t) = v(t) i(t) = V I [2 sin ωt sin(ωt − φ )] = V I[cos φ − cos(2ωt − φ )]

= V I cos φ (1 − cos 2ωt) −V I sin φ sin 2ωt

(2.2)

= P(1 − cos 2ωt) − Q sin 2ωt = pactive (t) + preactive (t)

The first term in above equation is called as active power, as it is the power that flows in the resistive element. The second term is similarly referred to as reactive power, as this component is present due to reactive elements such as inductance and capacitance.   Here, P = T1 0T pactive (t) dt = T1 0T p(t) dt is known by many names such as average active power, real power, useful power or simply power. In above T is the time period of voltage or current waveform. Therefore, real power can be expressed as, P = V I cos φ =

1 T

 T 0

pactive (t) dt =

1 T

 T 0

p(t) dt

(2.3)

Single-Phase Circuits: Power Definitions and Components

33

Reactive power, Q is defined as, ∆

max {preactive (t)} = V I sin φ

Q =

(2.4)

It should be noted that the way Q is defined is different from P. The term Q is defined as maximum value of the second term of (2.2) and not as an average value of the second term, although, it is denoted as average reactive power. This difference should be kept in mind while defining P and Q. Equation (2.2) shows that instantaneous power can be decomposed into two terms. The first term has an average value of V I cos φ and an alternating component of V I cos 2ωt, oscillating at twice the line frequency. This term, i.e., pactive is never negative and therefore is called unidirectional or dc power. The second term, i.e., preactive has an alternating component V I sin φ sin 2ωt oscillating at twice frequency with a peak value of V I sin φ . The second term has zero average value. Equation (2.2) can further be written in the following form: p(t) = V I cos φ −V I cos(2ωt − φ ) = =

p(t) ¯ + p(t) ˜ paverage + poscillating

(2.5)

With the above definitions of P and Q, the instantaneous power p(t) can be re-written as following. p(t) = P(1 − cos 2ωt) − Q sin 2ωt

(2.6)

Various components of single-phase instantaneous powers are shown in Fig. 2.2. As observed from the figure, the instantaneous power has positive and negative values forming a sinusoidal waveform of twice the fundamental frequency. The positive power is formed when both the voltage and current, are positive or negative, while negative power is formed when one of them is negative. The negative power is shown by a hatched region. Positive power indicates that the power is pumped from the source to the load, while negative power indicates that the power is returned to the source. The sum of average values of the positive and negative powers in a

Figure 2.2 Various powers in single-phase system

34

Power Quality in Power Distribution Systems: Concepts and Applications

cycle is the average or real power delivered to the load. As per equation (2.2), the instantaneous power, p(t) has two components, pactive (t) and preactive (t), as shown in the figure. The average value of p(t) and the average value of pactive (t) are same and equal to P = V I cos φ . Also, the maximum value of preactive (t) is reactive power Q = V I sin φ . These powers are indicated in the figure by dotted lines. The scalar apparent power, S in single-phase system is defined as the product of rms values of voltage and current as expressed below. S =V I

(2.7)

The apparent power, S can also be expressed in terms of real and reactive powers as given below.  V 2 I 2 cos2 φ +V 2 I 2 sin2 φ S =V I =  P2 + Q2 (2.8) = Therefore, we have

S2 = P2 + Q2 .

(2.9)

Another important term that indicates the usage of electric power is the power factor (p f ). The power factor in single-phase system with sinusoidal nature of voltage and current of the fundamental frequency is defined as the ratio of the real power to the apparent power. This is given below. Power factor = p f =

P S

(2.10)

Substituting the value of P from (2.3) and S from (2.7), the above equation can be simplified to, pf =

V I cos φ = cos φ VI

(2.11)

The above equation indicates that the power factor can also be defined as cosine angle between the voltage and current. However, this definition holds true only if the voltage and current are of the fundamental frequency. The relationship between P, Q, S and angle φ is represented using power triangle shown in Fig. 2.3.

Figure 2.3 Relation between P, Q, S, and φ

Single-Phase Circuits: Power Definitions and Components

2.3

35

PHASOR REPRESENTATION OF ELECTRICAL QUANTITIES

To analyze electrical circuits, we often use phasor notation to represent electrical quantities such as voltage, and current. To compute circuit parameters at different points in the circuit, we make use of voltage and current phasors as well as impedance and apparent power in complex form. In phasor notation, the voltage and currents are represented by a phasor having a magnitude of rms value and phase angle with respect to a reference phasor. A reference phasor is one whose angle is zero. Thus, the voltage and current expressed by (2.1), can be written in phasor form as following. V = V ∠0◦ I = I∠−φ

(2.12)

In the above equation, V , I are rms value of voltage and current, respectively. The angle φ is an angle between voltage and current. The voltage and current phasors are shown in Fig. 2.4. In case, we have an electrical load represented by linear resistance R, inductance L, and capacitance C parameters, the relationship between voltage and current can be expressed as below:

Figure 2.4 Voltage and current phasors

I=

V Z

(2.13)

where Z is a complex quantity and is known as the impedance of the electrical load. For RLC circuits, it has three components, i.e., resistance R, inductive reactance XL of inductance L at fundamental frequency and capacitive reactance XC of capacitance C at fundamental frequency. Further, for series connected RLC load, the impedance can be expressed as, Z = R + jXL − jXC .

(2.14)

Similarly, for RLC parallel connected load, the relationship between admittance, impedance can be expressed as following. Y=

1 1 1 1 = + + . Z R jXL − jXC

(2.15)

36

Power Quality in Power Distribution Systems: Concepts and Applications

In the above equations, XL = 2 π f L 1 XC = 2π f C

(2.16)

Where f is the frequency of supply voltage and ω = 2 π f . With the above description, complex power S can be expressed as below. S =V I



(2.17)



The term I refers to complex conjugate of I phasor. This operation ensures that the phase angle difference between the voltage and current is considered to compute complex power S. Substituting V and I from (2.12), we get, S

= V ∠0◦ I ∠φ = V I ∠φ = V I cos φ + jV I sin φ = P + j Q

(2.18)

From the above, the magnitude of the complex power, S is the apparent power, S as described in (2.8). That is,  |S| = S = P2 + Q2 (2.19) This gives similar relation as given in (2.9).

√ Example 2.1. Consider a sinusoidal supply voltage v(t) = 230 2 sin ωt V, supplying a linear load of impedance ZL = 12 + j13 Ω at ω = 2π f rad/s, f = 50 Hz. Write the expression for the current, i(t) in the circuit. Based on v(t) and i(t), determine the following. (a) Instantaneous power p(t), instantaneous active power pactive (t) and instantaneous reactive power preactive (t). (b) Compute average real power P, reactive power Q, apparent power S, and power factor p f . (c) Repeat the above when load is ZL = 12 − j13 Ω, ZL = 12 Ω, and ZL = j13 Ω (d) Comment upon the results. Solution: A single-phase circuit supplying linear load is shown in Fig. 2.1. In general, the current in the circuit is given as, √ i(t) = 2 I sin(ωt − φ ) where φ = tan−1 (XL /RL ), and I = V /|ZL |

Single-Phase Circuits: Power Definitions and Components

37

Case 1: When ZL = 12 + j13 Ω load is inductive, √ 2 2 |ZL | = RL + XL = 122 + 132 = 17.692 Ω, and I = 230/17.692 = 13 A φ = tan−1 (X/R) = tan−1 (13/12) = 47.29◦ Therefore, we have √ v(t) = 230 2 sin ωt √ i(t) = 13 2 sin(ωt − 47.29◦ ) The instantaneous power is given as, p(t)

= V I cos φ (1 − cos 2ωt) −V I sin φ sin 2ωt = 230 × 13 cos 47.29◦ (1 − cos 2ωt) − 230 × 13 sin 47.29◦ sin 2ωt

= =

2028.23(1 − cos 2ωt) − 2196.9 sin 2ωt pactive (t) + preactive (t)

The above implies that, pactive (t) = 2028.23(1 − cos 2ωt)

preactive (t) = −2196.9 sin 2ωt

Figure 2.5 Case 1: Voltage, current, and various power components

38

Power Quality in Power Distribution Systems: Concepts and Applications

Average real power (P) is given as, P P



1 T p(t) dt T 0 = V I cos φ = 230 × 13 × cos 47.29◦ = 2028.23 W =

Reactive power (Q) is given as maximum value of preactive , and equals to V I sin φ as given below. Q Apparant power, S Power factor, p f

= V I sin φ = 230 × 13 × sin 47.2906◦ = 2196.9 VAr  = V I = P2 + Q2 = 230 × 13 = 2990 VA P 2028.23 = = = 0.6783 (lag) S 2990

For this case, the voltage, current and various components of the power are shown in Fig. 2.5. As seen from the figure, the current lags the voltage due to inductive load. The pactive has an offset of 2028.23 W, which is also indicated as P in the right bottom graph. The preactive has zero average value and its maximum value is equal to Q, which is 2196.9 VArs. √ Case 2: When load is capacitive, ZL = 12 − j13 that implies |ZL | = 122 + 132 = 17.692 Ω, and I = 230/17.692 = 13 A, φ = tan−1 (−13/12) = −47.29◦ . √ v(t) = 230 2 sin ωt √ i(t) = 13 2 sin(ωt + 47.29◦ ) p(t) = V I cos φ (1 − cos 2ωt) −V I sin φ sin 2ωt = 230 × 13 cos(−47.29◦ )(1 − cos 2ωt) − 230 × 13 sin(−47.29◦ ) sin 2ωt = 2028.23(1 − cos 2ωt) + 2196.9 sin 2ωt

From the above result, pactive (t) = preactive (t) =

2028.23(1 − cos 2ωt) 2196.9 sin 2ωt

Average active, reactive and apparent powers are calculated as, P Q S Power factor

= V I cos φ = 230 × 13 × cos 47.29◦ = 2028.23 W

= V I sin φ = 230 × 13 × sin(−47.29◦ ) = −2196.9 VAr  = V I = P2 + Q2 = 230 × 13 = 2990 VA P 2028.23 = = = 0.6783 (lead) S 2990

For Case 2, the voltage, current and various components of the power are shown in Fig. 2.6. The explanation given earlier also holds true for this case.

Single-Phase Circuits: Power Definitions and Components

39

Figure 2.6 Case 2: Voltage, current, and various power components

Case 3: When load is resistive, ZL = RL = 12 Ω, I = 230/12 = 19.167 A, and φ = 0◦ . Therefore, we have √ v(t) = 230 2 sin ωt √ i(t) = 19.167 2 sin ωt p(t) = 230 × 19.167 cos 0◦ (1 − cos 2ωt) − 230 × 19.167 sin 0◦ sin 2ωt = 4408.33(1 − cos 2ωt)

From the above result, pactive (t) = 4408.33(1 − cos 2ωt) preactive (t) = 0 Average active, reactive and apparent powers are calculated as, P = V I cos φ = 230 × 19.167 × cos 0◦ = 4408.33 W Q = V I sin φ = 230 × 19.167 × sin 0◦ = 0 VAr  S = V I = P2 + Q2 = 230 × 19.167 = 4408.33 VA 4408.33 Power factor = =1 4408.33 For Case 3, the voltage, current and various components of the power are shown in Fig. 2.7. Since the load is resistive, as seen from the graph preactive

40

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 2.7 Case 3: Voltage, current, and various power components

is zero and p(t) is equal to pactive . The average value of p(t) is real power (P), which is equal to 4408.33 W. Case 4: When the load is purely reactive, ZL = j13 Ω , |ZL | = 13 Ω, I = 230 13 = 17.692 A, and φ = 90◦ . Therefore, we have √ v(t) = 230 2 sin ωt √ i(t) = 17.692 2 sin(ωt − 90◦ ) p(t) = 230 × 17.692 cos 90◦ (1 − cos 2ωt) − 230 × 17.692 sin 90◦ sin 2ωt = −4069.16 sin 2ωt

From the above result, pactive (t) = preactive (t) =

0 −4069.16 sin 2ωt

Average active, reactive and apparent powers are calculated as, P = V I cos φ = 230 × 17.692 × cos 90◦ = 0 W Q = V I sin φ = 230 × 17.692 × sin 90◦ = 4069.16 VAr  S = V I = P2 + Q2 = 230 × 17.692 = 4069.16 VA

Single-Phase Circuits: Power Definitions and Components

41

Figure 2.8 Case 4: Voltage, current, and various power components

Power factor =

0 =0 4069.16

For Case 4, the voltage, current and various components of the power are shown in Fig. 2.8. The load in this case is purely reactive, hence there is no average component of p(t). The maximum value of p(t) is same as preactive (t) or Q, which is equal to 4069 VArs.

2.4

SINUSOIDAL VOLTAGE SOURCE SUPPLYING NON-LINEAR LOAD CURRENT

Now a nonlinear load is considered, supplied by a sinusoidal voltage source. A nonlinear load is the one which consists of switched elements such as diode, transistors, IGBTs (Insulated Gate Bipolar Transistors) MOSFETs (Metal Oxide Semiconductor Field Effect Transistors), etc., in the circuit. In power circuit, non-linear load current exists, when source supplies to power electronics based loads such as rectifier, inverter, cyclo-converters, etc. These loads cause harmonics in the load current. Assuming a general case where all harmonics are present in the load current, the voltage and current are expressed as following. √ v(t) = 2V sin ωt √ ∞ i(t) = 2 ∑ In sin(nωt − φn ) (2.20) n=1

42

Power Quality in Power Distribution Systems: Concepts and Applications

where In and φn in the above equation represent the rms value and phase angle of nth harmonic component of current. The instantaneous power is therefore given by, p(t) = v(t) i(t) =



= V

2V sin ωt





2 ∑ In sin(nωt − φn ) n=1



∑ [In 2 sin ωt sin(nωt − φn )]

n=1

= V [I1 2 sin ωt sin(ωt − φ1 )] ∞

+V

∑ [In 2 sin ωt sin(nωt − φn )]

(2.21)

n=2

Note that, 2 sin A sin B = cos(A−B) −cos(A+B), using this, (2.21) can be re-written as the following. p(t)

= V I1 [cos φ1 − cos(2ωt − φ1 )] ∞

+V

∑ In {cos[(n − 1)ω t − φn ] − cos[(n + 1)ω t − φn ]}

n=2

= V I1 cos φ1 (1 − cos 2ωt) −V I1 sin φ1 sin 2ωt

(2.22)



+ ∑ V In {cos[(n − 1)ω t − φn ] − cos[(n + 1)ω t − φn ]} n=2

In above equation, average active power P and reactive power Q are given by, 1 T

 T

p(t) dt = V I1 cos φ1

P

= P1 = average value of p(t) =

Q

= Q1 = peak value of second term in (2.22) = V I1 sin φ1



0

(2.23) (2.24)

The apparent power S is given by S S

= V I = V I12 + I22 + I32 + .....

(2.25)

Equation (2.25) can be re-arranged as given below. S2

= V 2 I12 +V 2 (I22 + I32 + I42 + ...) = (V I1 cos φ1 )2 + (V I1 sin φ1 )2 +V 2 (I22 + I32 + I42 + .....)

= P2 + Q2 + H 2

(2.26)

In above equation, H is known as harmonic power and represents Volt-Ampere (VAs) rating corresponding to harmonics and is equal to,  H = V I22 + I32 + I42 + ..... (2.27) The following points are observed from the description.

Single-Phase Circuits: Power Definitions and Components

43

1. P and Q are dependent on the fundamental current components. 2. H is dependent on the current harmonic components. 3. Power components V I1 cos φ1 cos 2ωt and V I1 sin φ1 sin 2ωt are oscillating components with twice the fundamental frequency. The reactive component of power, i.e., V I1 sin φ1 sin 2ωt can be eliminated using appropriately chosen reactive network, such as capacitor or inductor. 4. There are other terms in (2.22), which are functions of multiple integer of fundamental frequency. These terms can be eliminated using tuned LC filters. Displacement Power Factor (DPF) or Fundamental Power Factor (p f1 ) is denoted by cos φ1 and is cosine angle between the fundamental voltage and current. This is equal to, DPF = p f1 = cos φ1 =

P1 . S1

(2.28)

The Power Factor (p f ) is defined as ratio of real power to the total apparent power (V I) and is expressed as, Power Factor = p f = cos φ

P S   V I1 cos φ1 I1 = = cos φ1 VI I = cos γ cos φ1 =

(2.29)

Equation (2.29) shows that the power factor becomes less by a factor of cos γ, which is the ratio of rms value of the fundamental to the rms value of the total current. This is due to the presence of harmonics in the load current. The nonlinear load current increases the ampere rating of the conductor for the same amount of real power transfer, and therefore such types of loads are not desirable in a power system. The relationship between the real power (P), reactive power (Q), harmonic power (H), DPF angle (φ1 ), and power factor angle (φ ) is depicted in Fig. 2.9. This is also known as power tetrahedron.

Figure 2.9 Power tetrahedron showing relation among P, Q, H, S, φ , and φ1

44

Power Quality in Power Distribution Systems: Concepts and Applications

Example 2.2. Consider an ideal single-phase voltage source √ supplying a rectifier load as given in Fig. 2.10. Given a supply voltage, vs (t) = 230 2 sin ωt and source impedance is negligible. The load inductance is quite large so that the output current is constant. Draw the voltage and current waveforms. Express the source current using Fourier series. Based on that determine the following. (a) Plot instantaneous power p(t). (b) Plot components of p(t) i.e., pactive (t), preactive (t). (c) Compute average real power, reactive power, apparent power, power factor, displacement power factor (or fundamental power factor). (d) Comment upon the results in terms of VA rating and power output.

Figure 2.10 A single-phase system with non-linear load

Solution: The supply voltage and current are shown in Fig. 2.11. The current waveform is of the square type and its Fourier series expansion is given below [19]–[22]. is (t) =



4Id sin hωt h=2n+1 hπ



where n = 0, 1, 2 . . .

The instantaneous power is therefore given by, p(t) = vs (t) is (t) =



2V sin ωt



4Id sin hωt. h=2n+1 hπ



By expansion of the above equation, the average active power (P) and reactive power (Q) are given as below. P

= P1 = average value of pactive (t) or p(t) = V I1 cos φ1 (since, φ1 = 0, cos φ1 = 1, sin φ1 = 0)

= V I1 Q



= Q1 = peak value of preactive (t) = V I1 sin φ1 = 0

Single-Phase Circuits: Power Definitions and Components

45

Figure 2.11 Supply voltage, current, and power waveforms

The dc side inductance is relatively large to make output dc current (Id ) as a constant value. This is also clear from the current waveform obtained from the simulation. From the waveform, it is found that Id = 103.5 A. Alternatively, this current can be found by the following expression. √ √ 2 2 230 1 2 2V 1 = = 103.5 A Id = Vd /R = π R π 2 In the above expression, Vd is the average value of rectifier output voltage, vd . The rms values of total and fundamental source current are given below. Isrms Is1

= Id = 103.5 A √ 2 2 = Id = 93.15 A π

(c) The real power (P) is given by P

= V Is1 cos φ1 = V Is1 cos 0◦ = V Is1 √ 2 2 Id = 21424.5 W = V× π

The reactive power (Q) is given by Q = Q1 = V Is1 sin φ1 = V Is1 sin 0◦ = 0

46

Power Quality in Power Distribution Systems: Concepts and Applications

The apparent power (S) is given by S

= V Isrms = V Id = 23805 VA

The displacement power factor (cos φ1 ) is, = cos φ1 = 1

DPF Therefore power factor is given by, pf

= cos φ = = =

P P1 = S S

V Is1 cos φ1 V Isrms Is1 × DPF = 0.9 lag Isrms

(d) As seen from the results that the overall power factor reduces due to the presence of harmonics in the load current. This has effect of increasing the VA rating of the system. Example 2.3. Compute the following for the source current of the system shown in Fig. 2.10 with RL = 15 Ω. Assume that the load inductor is very high as to cause constant dc current (Id ) at the output terminals and source voltage has negligible impedance. (a) Power factor, displacement factor. (b) Total harmonic distortion. (c) Input real and reactive powers, output power. Solution: (a) The average voltage at the dc link (Vd ) is given by, √ 2 × 230 2 2Vm = = 207.07 V Vd = π π Therefore, the average dc current is computed as follows. Id =

2Vm = 13.84 A π RL

The source current can be expressed as,   sin 3ωt sin 5ωt 4Id sin ωt + + + ... is (t) = π 3 5 sin 5ωt sin 3ωt is (t) = 17.62 sin ωt + 5.87 + 3.52 + ... 3 5

Single-Phase Circuits: Power Definitions and Components

47

Figure 2.12 Supply voltage and current waveforms

The corresponding voltage and current waveforms are shown in Fig. 2.12 The rms value of nth harmonic component is given as, Isn =

4Id √ nπ 2

The rms value of the fundamental component of the source current is given as, Is1 =

4 × 13.84 √ = 12.45 A π 2

Displacement factor can be computed as, cos φ1 = cos 0◦ = 1.0 (b) The power factor is given as, pf =

Vs1 Is1 cos φ1 230 × 12.45 × 1 = = 0.9 lag. VI 230 × 13.84

Total harmonic distortion (T HD) of the source current is given as,  √ 2 Is2 − Is1 13.842 − 12.452 × 100 = × 100 = 48.55% T HD% = Is1 12.45 (c) The input real power (Ps ) is given by Ps = V Is1 = 230 × 12.45 = 2863.5 W The input reactive power (Qs ) is given by Qs = Q1 = 0 VAr The output power (Po ) is given by Po = Vd Id = 207.07 × 13.84 = 2857.56 W.

48

Power Quality in Power Distribution Systems: Concepts and Applications

2.5

NON-SINUSOIDAL VOLTAGE SOURCE SUPPLYING LINEAR AND NON-LINEAR LOADS

The voltage source too may have harmonics transmitted from generation or produced due to nonlinear loads in presence of feeder impedance. In this case, we shall consider generalized case of non-sinusoidal voltage source supplying nonlinear loads including dc components. For this system, the voltage and current are represented as, ∞

v(t) = Vdc + ∑



n=1

2Vn sin(nωt − φvn )

(2.30)

2In sin(nωt − φin )

(2.31)

and, ∞

i(t) = Idc + ∑



n=1

In the above equations, Vn , In , φvn , φin are rms magnitude and phase angle of nth harmonic component of voltage and current, respectively. The instantaneous power p(t) is given by,    ∞ √ ∞ √ p(t) = Vdc + ∑ 2Vn sin(nωt − φvn ) Idc + ∑ 2In sin(nωt − φin ) (2.32) n=1

p(t)

n=1

∞ √ ∞ √ = Vdc Idc +Vdc ∑ 2In sin(nωt − φin ) + Idc ∑ 2Vn sin(nωt − φvn )    n=1 n=1 I       ∞

+∑

II



n=1



2Vn sin(nωt − φvn )





n=1



III



2In sin(nωt − φin )

IV

p(t) = pdc−dc + pdc−ac + pac−dc + pac−ac



(2.33)

(2.34)

The term I (pdc−dc ) contributes to power from dc components of voltage and current. Terms II (pdc−ac ) and III (pac−dc ) result from the interaction of dc and ac components of voltage and current, respectively. In case, there are no dc components all these power components are zero. In practical cases, dc components are very less and the first three terms have negligible value compared to IV term. Thus, we shall focus on IV (pac−ac ) term which corresponds to ac components present in power system. The IVth term can be written as,    ∞ √ ∞ √ pac−ac = ∑ 2Vn sin(nωt − φvn ) (2.35) ∑ 2Ih sin(hωt − φih ) n=1

h=1

49

Single-Phase Circuits: Power Definitions and Components

where n = h = 1, 2, 3..., similar frequency terms will interact. When n = h, dissimilar frequency terms will interact. This is expressed below. √ √ pac−ac (t) = 2V1 sin(ωt − φv1 ) 2I1 sin(ωt − φi1 )    A





+ 2V1 sin(ωt − φv1 )





h=2,h=1

2Ih sin(hωt − φih )

 B √ + 2V2 sin(2ωt − φv2 ) 2I2 sin(2ωt − φi2 )     √

A





+ 2V2 sin(2ωt − φv2 )



h=1,h=2

√ 2Ih sin(hωt − φih ) + . . . + . . .

  B √ √ + 2Vn sin(nωt − φvn ) 2In sin(nωt − φin )    A





+ 2Vn sin(nωt − φvn ) 







h=1,h=n





2Ih sin(hωt − φih )

(2.36)



B

The terms in A of above equation form similar frequency terms and terms in B form dissimilar frequency terms. We shall denote these terms by pac−ac−nn and pac−ac−nh , respectively, which are given below. pac−ac−nn (t) =

pac−ac−nh (t) =







n=1



∑ 2Vn In sin(nωt − φvn ) sin(nωt − φin )

(2.37)

n=1

2Vn sin(nωt − φvn )





h=1,h=n

√ 2In sin(hωt − φih )

(2.38)

Now, let us simplify pac−ac−nn term in (2.37) pac−ac−nn (t) = =



∑ Vn In [cos(φin − φvn ) − cos(2nωt − φin − φvn )]

n=1 ∞

∑ Vn In [cos φn − cos(2nωt − (φin − φvn ) − 2φvn )]

n=1 ∞

=

∑ Vn In [cos φn − cos 2(nωt − φvn ) − φn ]

n=1 ∞

=

∑ Vn In [cos φn − cos 2(nωt − φvn ) cos φn

n=1

− sin 2(nωt − φvn ) sin φn ]

(2.39)

50

Power Quality in Power Distribution Systems: Concepts and Applications

where φn = (φin − φvn ) is the phase angle between nth harmonic components of current and voltage. Therefore, ∞

∑ Vn In cos φn {1 − cos 2(nωt − φvn )}

pac−ac−nn (t) =

n=1 ∞

− ∑ Vn In sin φn sin 2(nωt − φvn ).

(2.40)

n=1

Thus, the instantaneous power is given by, p(t)

p(t)

=

+p +p +p +p p    ac−ac−nh  dc−dc    dc−ac    ac−dc    ac−ac−nn   IVA I II III IVB   

= Vdc Idc +Vdc ∞

IV







n=1

2In sin(nωt − φin ) + Idc





n=1



2Vn sin(nωt − φvn )

+ ∑ Vn In cos φn [1 − cos 2(nωt − φvn )] n=1 ∞

− ∑ Vn In sin φn sin 2(nωt − φvn ) n=1

+ 2.5.1









n=1

2Vn sin(nωt − φvn )







h=1, h=n



2Ih sin(hωt − φih )



(2.41)

ACTIVE POWER

Instantaneous active power, pactive (t) in (2.40), is expressed as, ∞

pactive (t) = Vdc Idc + ∑ Vn In cos φn [1 − cos 2(nωt − φvn )]

(2.42)

n=1

It has non-negative value with some average component, giving average active power. Therefore, P

=

1 T

 T 0

p(t) dt = ∞

1 T

 T 0

v(t) i(t) dt

= Vdc Idc + ∑ Vn In cos φn

(2.43)

n=1

The reactive component of the instantaneous power is denoted by preactive (t) and is given as follows. ∞

preactive (t) = − ∑ Vn In sin φn sin 2(nωt − φvn ) n=1

(2.44)

Single-Phase Circuits: Power Definitions and Components

51

resulting in Q 

sum of maximum value of each term in (2.44) ∞

∑ Vn In sin φn .

=

(2.45)

n=1

From (2.43) P =



Pdc + ∑ Vn In cos φn n=1

= Pdc +V1 I1 cos φ1 +V2 I2 cos φ2 +V3 I3 cos φ3 + . . . = Pdc + P1 + P2 + P3 + . . . = Pdc + P1 + PH

(2.46)

In the above equation, Pdc = Average active power corresponding to the dc components P1 = Average fundamental active power PH = Average harmonic active power Average fundamental active power (P1 ) can also be found from fundamentals of voltage and current, i.e., P1

1 T

=

 T 0

v1 (t) i1 (t)dt

(2.47)

and harmonic active power (PH ) can be found as below. PH 2.5.2



=

∑ Vn In cos φn = P − P1 .

(2.48)

n=2

REACTIVE POWER

The reactive power (Q), can be found by summing the maximum value of each term in (2.44). This is given below. Q =



∑ Vn In sin φn

n=1

= V1 I1 sin φ1 +V2 I2 sin φ2 +V3 I3 sin φ3 + . . . = Q1 + Q2 + Q3 + . . . = Q1 + QH

(2.49)

Usually, this reactive power is referred to as Budeanu’s reactive power and we use subscript “B,” to indicate this, i.e., QB = Q1B + QHB

(2.50)

52

Power Quality in Power Distribution Systems: Concepts and Applications

The remaining dissimilar frequency terms of (2.41) are accounted using prest (t). Therefore, we can write, p(t) = pdc−dc + pactive (t) + preactive (t) +    Similar frequency terms

where,

pdc−dc

(2.51)

Dissimilar frequency terms

= Vdc Idc ∞

pactive (t)

=

preactive (t)

=

prest (t)

prest (t)   

∑ Vn In cos φn {1 − cos 2(nωt − φvn )}

n=1 ∞

− ∑ Vn In sin φn sin 2(nωt − φvn ) n=1 ∞

= Vdc

√ ∑ 2In sin(nωt − φin ) + Idc

n=1 ∞ √

+∑

n=1

2Vn sin(nωt − φvn )









n=1





h=1, h=n

2Vn sin(nωt − φvn )

2Ih sin(mωt − φih ) (2.52)

2.5.3

APPARENT POWER

The scalar apparent power which is defined as product of rms value of voltage and current, is expressed as follows. S

where,

= V I  2 +V 2 +V 2 + · · · 2 + I2 + I2 + · · · = Vdc Idc 1 2 1 2   2 +V 2 +V 2 2 + I2 + I2 Vdc Idc = H H 1 1 VH2

= V22 +V32 + · · · =

IH2

= I22 + I32 + · · · =

(2.53)



∑ Vn2

n=2 ∞ In2 n=2



(2.54)

VH and IH are denoted as harmonic voltage and harmonic current, respectively. Expanding (2.53) we can write S2

= V 2I2 2 2 = (Vdc +V12 +VH2 )(Idc + I12 + IH2 ) 2 2 2 2 2 2 2 2 = Vdc Idc +Vdc I1 +Vdc IH +V12 I12 +V12 Idc +V12 IH2 +VH2 Idc +VH2 I12 +VH2 IH2 2 2 2 2 2 = Vdc Idc +V12 I12 +VH2 IH2 +Vdc (I1 + IH2 ) + Idc (V12 +VH2 ) +V12 IH2 +VH2 I12

Single-Phase Circuits: Power Definitions and Components

= = =

53

2 2 2 Sdc + S12 + SH + SD

2 2 2 S12 + Sdc + SH + SD   

S12 + SN2

(2.55)

In the above equation, the terms, S1 , Sdc , SH , and SD are fundamental, dc, harmonic, and distortion (due to dissimilar voltage and current terms) apparent powers, respectively. The term, SN is known as non-fundamental apparent power and is given as follows. 2 2 2 2 2 2 2 2 2 2 SN2 = Vdc I1 +Vdc IH +V12 Idc +V12 IH2 +VH2 Idc +VH2 I12 +VH2 IH2 + Idc IH + Idc Vdc (2.56)

Practically in power systems dc components are negligible. Therefore neglecting the contribution of Vdc and Idc associated terms in (2.56), the following is obtained. SN2

I12VH2 +V12 IH2 +VH2 IH2

=

2 DV2 + D2I + SH

=

(2.57)

The terms DI and DV in (2.57) are known as apparent powers due to distortion in current and voltage, respectively. These are given below. DV

=

DI

= V1 IH

I1 VH (2.58)

These are further expressed in terms of THD components of voltage and current, as given below. T HDV

=

T HDI

=

VH V1 IH I1

(2.59)

From (2.59), the harmonic components of current and voltage are expressed below. VH

= T HDV V1

IH

= T HDI I1

(2.60)

Using (2.58) and (2.60), DV

= V1 I1 T HDV = S1 T HDV

DI

= V1 I1 T HDI = S1 T HDI

SH

= VH IH = S1 T HDI T HDV

(2.61)

Therefore using (2.57) and (2.61), SN could be expressed as following. SN2 = S12 (T HD2I + T HDV2 + T HD2I T HDV2 )

(2.62)

54

Power Quality in Power Distribution Systems: Concepts and Applications

Normally in power system, T HDV > PH and T HDV 0 2

(4.105)

Similar to (4.103), we can express phase-b current in terms of Im (α I bl ), as given below. √ √ ibl (t) = 2 Im(I bl e jωt ) = 2 Im (α I bl e jωt α −1 ) √   = 2 Im α I bl [cos(ωt − 120◦ ) + j sin(ωt − 120◦ )] (4.106) From the above equation, we get the following. ibl (t) Im (αI bl ) = √ 2

at ωt − 120◦ = 0, which implies vb (t) = 0, dvb /dt > 0 (4.107)

Theory of Fundamental Load Compensation

173

Similarly, for phase-c, it can be proved that, icl (t) Im (α 2 I cl ) = √ 2

when, vc (t) = 0, dvc /dt > 0

(4.108)

Substituting Im (I al ) , Im (αI bl ) and Im (α 2 I cl ) from (4.105), (4.107), and (4.108), respectively, in (4.102), we get the following.   1 √ Bab = − | + i | − i | i dva al bl (v =0, dvb >0) cl (vc =0, dvc >0) γ dt b dt 3 2V  (va =0, dt >0)  1 √ = − | + i | + i | −i Bbc dv dv dv al (va =0, a >0) bl (v =0, b >0) cl (vc =0, c >0) (4.109) γ dt dt b dt 3 2V   1 ca + icl |(v =0, dvc >0) ial |(v =0, dva >0) − ibl | Bγ = − √ dv a c (vb =0, dtb >0) dt dt 3 2V Thus, the desired compensating susceptances are expressed in terms of the three line currents sampled at instants defined at positive-going zero crossings of the line to neutral voltages va , vb , vc . An artificial neutral at ground potential may be created by measuring voltages va , vb , and vc to implement above algorithm. Since the method involves computation of samples of load currents, it is called as sampling method. 4.5.2

AVERAGING METHOD

In this method, we express the compensator susceptances in terms of real and reactive power terms and finally expressed them in time domain through averaging process. The method is described below. From (4.102), susceptance, Bab γ , can be re-written as following.     1    Im I al + Im αI bl − Im α 2 I cl 3V       1  = − 2 V Im I al +V Im αI bl −V Im α 2 I cl 3V      1   = − 2 Im V I al + Im V αI bl − Im V α 2 I cl 3V

Bab γ =−

(4.110)

Note the following property of phasors and applying it for the simplification of the above expression.  ∗  ∗ ∗   (4.111) Im V I = −Im V I = −Im V I Using above equation, (4.110) can be written as,

 1  Im (V a I al )∗ + Im (αV a I bl )∗ − Im (α 2V I cl )∗ 2 3V 1  ∗ ∗ ∗ ∗ ∗ ∗  Im (V a I al ) + Im (α ∗V a I bl ) − Im ((α 2 )∗V a I cl ) = 2 3V 1  ∗ ∗ ∗ ∗ ∗ ∗  = Im (V a I al ) + Im (α 2V a I bl ) − Im (αV a I cl ) 2 3V

Bab γ =

174

Power Quality in Power Distribution Systems: Concepts and Applications ∗



Since V a = V ∠0◦ is a reference phasor, therefore V a = V a = V , α 2 V a = V b and ∗ α V a = V c . Using this, the above equation can be written as follows. 1  ∗ ∗ ∗  Im (V a I al ) + Im (V b I bl ) − Im (V c I cl ) 2 3V 1  ∗ ∗ ∗  bc Similarly, Bγ = −Im (V a I al ) + Im (V b I bl ) + Im (V c I cl ) (4.112) 3V 2 1  ∗ ∗ ∗  Bca Im (V a I al ) − Im (V b I bl ) + Im (V c I cl ) γ = 3V 2 Bab γ =

It can be further proved that,



1 T va (t)∠(−π/2) ial (t) dt T 0  T 1 ∗ Im (V b I b ) = vb (t)∠(−π/2) ibl (t) dt T 0  1 T ∗ Im (V c I c ) = vc (t)∠(−π/2) icl (t) dt T 0 ∗

Im (V a I a ) =

(4.113)

In (4.113), the term va (t)∠(−π/2) denotes the voltage va (t) shifted by −π/2 radian in time domain. For fundamental balanced voltages, the following relationship between phase and line voltages are true. √ √ va (t)∠(−π/2) = 2V sin(ωt − π/2) = vbc (t)/ 3 √ √ vb (t)∠(−π/2) = 2V sin(ωt − 2π/3 − π/2) = vca (t)/ 3 (4.114) √ √ vc (t)∠(−π/2) = 2V sin(ωt + 2π/3 − π/2) = vca (t)/ 3 From (4.113) and (4.114), the following can be written. 

T 1 ∗ Im (V a I a ) = √ vbc (t) ial (t) dt 3T 0  T 1 ∗ Im (V b I b ) = √ vca (t) ibl (t) dt 3T 0  T 1 ∗ Im (V c I c ) = √ vab (t) icl (t) dt 3T 0

(4.115)

In the above, T is the time period of system voltage. Substituting the above values of ∗ ∗ ∗ Im (V a I a ), Im (V b I b ), and Im (V c I c ) into (4.112), we get the following. Bab γ

=

Bbc γ

=

Bca γ

=



1 T 1 √ (vbc ial + vca ibl − vab icl ) dt (3 3V 2 ) T 0  1 T 1 √ (−vbc ial + vca ibl + vab icl ) dt (3 3V 2 ) T 0 1 1 √ 2 (3 3V ) T

 T 0

(vbc ial − vca ibl + vab icl ) dt

(4.116)

175

Theory of Fundamental Load Compensation

The above equations can directly be used to know the compensator susceptances by performing averaging the product of the line to line voltages and phase load cur rents as per (4.116). The term 0T (.)dt = tt11−T (.)dt, where t1 is any arbitrary instant, can be implemented using moving an average of one cycle. This improves transient response by computing the average value at each instant. But in this case, the controller response, which changes the susceptance value, should match the speed of the computing algorithm.

4.6

COMPENSATOR ADMITTANCE REPRESENTED AS POSITIVE AND NEGATIVE SEQUENCE ADMITTANCE NETWORK

Recalling the following relations from equation (4.96) for unity power factor operation i.e., β = 0, we get the following. Bab γ

=

Bbc γ

=

Bca γ

=

√ −1 √ [Im (I 1l ) + Im (I 2l ) − 3 Re (I 2l )] 3 3V −1 √ [Im (I 1l ) − 2Im (I 2l )] 3 3V √ −1 √ [Im (I 1l ) + Im (I 2l ) + 3 Re (I 2l )] 3 3V

(4.117)

From these equations, it is evident that the first terms form the positive sequence susceptance as they involve I 1l terms. Similarly, the second and third terms in the above equation form negative sequence susceptance of the compensator, as these involve I 2l terms. Thus, we can write, Bab γ

=

ab Bab γ1 + Bγ2

Bbc γ

=

bc Bbc γ1 + Bγ2

Bca γ

(4.118)

ca Bca γ1 + Bγ2

=

Therefore, 1 bc ca [Im(I¯1l )] Bab γ1 = Bγ1 = Bγ1 = − √ 3 3V

(4.119)

and, Bab γ2 Bbc γ2 Bca γ2

1 = − √ 3 3V 1 = − √ 3 3V 1 = − √ 3 3V

 

 √ Im(I 2l ) − 3Re(I 2l ) 

−2 Im(I 2l )



 √ Im(I 2l ) + 3Re(I 2l )

(4.120)

176

Power Quality in Power Distribution Systems: Concepts and Applications

Earlier in (4.74), it was established that, I 0l I 1l I 2l Noting that,

= 0 √  = Ylab +Ylbc +Ylca 3V  √ = − α 2Ylab +Ylbc + αYlca 3V Ylab

ab = Gab l + jBl

Ylca

ca = Gca l + jBl

Ylbc

bc = Gbc l + jBl

Therefore, Im (I 1l ) = =

√ Im (Ylab +Ylbc +Ylca ) 3V √ bc ca (Bab l + Bl + Bl ) 3V

(4.121)

Thus, (4.119) is re-written as following. bc ca Bab γ1 = Bγ1 = Bγ1 = −

 1  ab ca Bl + Bab l + Bl 3

(4.122)

bc ca Now we shall compute Bab γ2 , Bγ2 , and Bγ2 using (4.120) as following. We know that,

√  3V I¯2l = − α 2Ylab +Ylbc + αYlca  √     −1 j 3 ab bc Gab + Gbc =− − l + jBl l + jBl 2 2   √  √ −1 j 3 ca + + (Gca 3V l + jBl ) 2 2  √ √ Gab Gca 3 ab 3 ca bc l l =− − + Bl + Gl − − B 2 2 2 2 l  √ √ √ 3 ab Bab 3 ca Bca bc l l −j 3V Gl + − Bl − Gl + 2 2 2 2  √ √ Gca Gab 3 ab 3 ca l l = − Bl − Gbc + B + l 2 2 2 2 l  √ √ ca √ B 3 ab Bab 3 l +j 3V Gl + l − Bbc Gca l − l + 2 2 2 2

(4.123)

177

Theory of Fundamental Load Compensation

The above implies that, √

 √ √ 3 ab Bab 3 ca Bca bc l l 3V Im (I 2l ) = Gl + − Bl − Gl + 2 2 2 2   √ √ √ 3 ab 3 ab √ bc 3 ca 3 ca √ − 3 Re(I 2l ) = − G + Bl + 3Gl − G − Bl 3V 2 l 2 2 l 2

Thus, Bab γ2 can be given as,

Similarly,

 √ 1  Im(I 2l ) − 3Re(I 2l ) Bab γ2 = − √ 3 3V √ bc √ ca  √ 1  ab ca 2Bl − Bbc − B + 3 Gl − 3 Gl 3V =− √ l l 3 3V  √  bc 1 bc ca ca = − 2Bab l − Bl − Bl + 3 Gl − Gl 3  1  1  ca ca ab Bbc + = √ Gl − Gbc l l + Bl − 2Bl 3 3

(4.124)

 √ 1  Bca Im(I¯2l ) + 3 Re(I¯2l ) γ2 = − √ 3 3V  √ √ 3 ab Bab 3 ca Bca 1 bc l Gl + − Bl − G + l =− √ 2 2 2 l 2 3 3V  √ √ 3 ab 3 ab √ bc 3 ca 3 ca √ G − Bl − 3Gl + G + Bl 3V + 2 l 2 2 l 2  1 √ ab √ bc bc ca 3Gl − 3Gl − Bab − B + 2B =− l l l 3  1  1  bc ab bc ca B = √ Gl − Gab + B − 2B + (4.125) l l l 3 l 3

and, Bbc γ2 is computed as below.

1 [−2Im(I¯2l )] Bbc γ2 = − √ 3 3V  √ √ √ 2 3 ab Bab 3 ca Bca bc l l V 3 = √ Gl + − Bl − Gl + 2 2 2 2 3 3V     1 ab 1 ca bc Bl + Bca + = √ Gab l − Gl l − 2Bl 3 3

(4.126)

178

Power Quality in Power Distribution Systems: Concepts and Applications

Using (4.119), (4.124)–(4.126), the overall compensator susceptances can be found as follows. ab ab Bab γ = Bγ1 + Bγ2    1  1  ca 1 bc ca bc ca bc ab √ B + + G = − Bab + B + B − G + B − 2B l l l l l l 3 l 3 l 3   1 bc = −Bab Gca l +√ l − Gl 3

Similarly,

and,

bc bc Bbc γ = Bγ1 + Bγ2   1  1  1 bc ca ca ca bc Bab + √ Gab + = − Bab l + Bl + Bl l − Gl l + Bl − 2Bl 3 3 3   1 ca = −Bbc Gab l +√ l − Gl 3

ca ca Bca γ = Bγ1 + Bγ2    1  1 1  bc bc ca ab bc ab ca √ B = − Bab + B + B − G + B − 2B + + G l l l l l l 3 l 3 l 3   1 ab = −Bca Gbc l +√ l − Gl 3

Thus, the compensator susceptances in terms of load parameters are given as follows.  1  ca ab Bab Gl − Gbc γ = −Bl + √ l 3   1 bc ca Bbc (4.127) Gab γ = −Bl + √ l − Gl 3   1 ca ab Bca Gbc γ = −Bl + √ l − Gl 3

It is interesting to observe the above equations. The first part of the equation nullifies the effect of the load susceptances and the second part compensates the effect of unbalance in the resistive parts of the load. The two terms together as compensator susceptance, make source currents balanced and in phase with the supply voltages. The compensator’s positive and negative sequence networks are shown in Fig. 4.14. The expressions in (4.127) now explain that earlier in Sub-section 4.4.4, why we √ √ ca = − jGab / 3 for load admittances, have taken, Yγab = 0, Yγbc = jGab / 3, and Y γ l l bc ca Ylab = Gab l , Yl = 0, and Yl = 0, to achieve balanced three-phase source currents.

Theory of Fundamental Load Compensation

179

What happens if we just use the following values of the compensator susceptances as given below? ab Bab γ = −Bl bc Bbc γ = −Bl ca Bca γ = −Bl

(4.128)

In the above case, load susceptance parts of the admittance are fully compensated. However, the source currents after compensation remain unbalanced due to unbalance resistive parts of the load. The compensator susceptances can be realized using lumped reactive components. When the load changes, their values have to be changed or switched to different values as calculated using above mentioned methods. This scheme is called an open-loop control scheme. The closed-loop control schemes can be realized using FACTs devices such as Fixed Capacitor-Thyristor Controlled Reactor (FC-TCR) [1], [12], [13], [15].

Figure 4.14 Sequence networks of the compensator

Example 4.5. For a delta connected load shown in Fig. 4.15, the load admittances are given as following, ab Ylab = Gab l + jBl bc Ylbc = Gbc l + jBl ca Ylca = Gca l + jBl

Given the load parameters: Zlab = 1/Ylab = 5 + j12 Ω Zlbc = 1/Ylbc = 3 + j4 Ω

Zlca = 1/Ylca = 9 − j13 Ω

180

Power Quality in Power Distribution Systems: Concepts and Applications

bc ca Determine compensator susceptances (Bab γ , Bγ , Bγ ) so that the supply sees the load as balanced with unity power factor. Also, find the line currents and source active and reactive powers before and after compensation.

Figure 4.15 A delta connected compensated load

Solution: Zlab = 5 + j12 Ω ⇒ Ylab = 0.03 − j0.0710  Zlbc = 3 + j4 Ω ⇒ Ylbc = 0.12 − j0.16 

Zlca = 9 − j12 Ω ⇒ Ylca = 0.04 + j0.0533  Once we know the admittances, we can compute line to line conductances and susceptances as given in the following. ab Gab l = 0.03 , Bl = −0.0710  bc Gbc l = 0.12 , Bl = −0.16 

ca Gca l = 0.04 , Bl = 0.0533 

1 ab ca bc Bab γ = −Bl + √ (Gl − Gl ) = 0.0248  3 1 bc ab ca Bbc γ = −Bl + √ (Gl − Gl ) = 0.1540  3 1 ca bc ab Bca γ = −Bl + √ (Gl − Gl ) = −0.0011  3 Total admittances are, Y ab = Ylab +Yγab = 0.03 − j0.0462  Y bc = Ylbc +Yγbc = 0.12 − j0.006 

Y ca = Ylca +Yγca = 0.04 + j0.0522 

Theory of Fundamental Load Compensation

181

Knowing these total admittances, we can find line currents as below. Current before Compensation   I al = I abl − I cal = (1 − α 2 )Ylab − (α − 1)Ylca V = 0.2150V ∠−9.51◦ A   I bl = I bcl − I abl = (α 2 − α)Ylbc − (1 − α 2 )Ylab V = 0.4035V ∠−161.66◦ A   I cl = I cal − I bcl = (α − 1)Ylca − (α 2 − α)Ylbc V = 0.2358V ∠43.54◦ A

Powers before compensation Sa Sb

= V a (I al )∗ = Pa + jQa = V × (0.2121 + j0.0355) = V b (I bl )∗ = Pb + jQb = V × (0.3014 + j0.2682)

Sc = V c (I cl )∗ = Pc + jQc = V × (0.0552 + j0.2293) Total real power, P = Pa + Pb + Pc = V × 0.5688W Total reactive power, Q = Qa + Qb + Qc = V × 0.5330 VAr

Power factor in phase-a, p fa

=

cos φa = cos(9.51◦ ) = 0.9863 lag

Power factor in phase-b, p fb

=

cos φb = cos(41.63◦ ) = 0.7471 lag

Power factor in phase-c, p fc

=

cos φc = cos(76.45◦ ) = 0.2334 lag

Thus, we observe that the phases draw reactive power from the lines and currents are unbalanced in magnitude and phase angles. Currents after compensation   I a = I ab − I ca = (1 − α 2 )Y ab − (α − 1)Y ca V = 0.1896 ×V ∠0◦ A   I b = I bc − I ab = (α 2 − α)Y bc − (1 − α 2 )Y ab V = 0.1896 ×V ∠−120◦ A   I a = I ca − I bc = (α − 1)Ylca − (α 2 − α)Y bc V = 0.1896 ×V ∠120◦ A

Powers after compensation

Sa

= V a (I a )∗ = Pa + jQa = V (0.1896 + j0.0)

Sb

= V b (I b )∗ = Pb + jQb = V (0.1896 + j0.0)

Sc = V c (I c )∗ = Pc + jQc = V (0.1896 + j0.0) Total real power, P = Pa + Pb + Pc = (V × 0.5688 )W Total reactive power, Q = Qa + Qb + Qc = 0 VAr Power factor in phase-a, p fa

=

cos φa = cos(0◦ ) = 1.0

Power factor in phase-b, p fb

=

cos φb = cos(0◦ ) = 1.0

Power factor in phase-c, p fc

=

cos φc = cos(0◦ ) = 1.0

From the above results, we observe that after placing the compensator of suitable values as calculated above, the line currents become balanced and have a unity power factor relationship with their respective voltages.

182

Power Quality in Power Distribution Systems: Concepts and Applications

Example 4.6. A three-phase system has delta connected load with Zab = 10+ j15 Ω, Zbc = 15+ j20 Ω and Zca = 10− j15 Ω supplied by balanced phase voltages with rms value of 230 V at 50 Hz. bc ca (a) Compute these susceptances (Bab γ , Bγ , and Bγ ) and express them in terms of physical parameters i.e., inductance or capacitance and their values.

(b) Determine the source currents after placing compensator. (c) Determine the source, compensator load active and reactive powers. (d) Draw the circuits as seen from the load and from the source side. Solution: (a) Compensator susceptances The load admittances are computed as follows. Zlab = 10 + j15 Ω ⇒ Ylab = 0.0308 − j 0.0462  Zlbc = 15 + j20 Ω ⇒ Ylbc = 0.0240 − j 0.0320 

Zlca = 10 − j15 Ω ⇒ Ylca = 0.0308 + j 0.0462 

Once we know the admittances we know,

ab Gab l = 0.0308 , Bl = −0.0462  bc Gbc l = 0.0240 , Bl = −0.0320  ca Gca l = 0.0308 , Bl = 0.0462 

1 ab ca bc Bab γ = −Bl + √ (Gl − Gl ) = 0.0501  3 1 bc ab ca Bbc γ = −Bl + √ (Gl − Gl ) = 0.0320  3 1 ca bc ab Bca γ = −Bl + √ (Gl − Gl ) = −0.0501  3 The values of physical parameters are computed as follows. Bab γ Bbc γ Bca γ

0.0501 = 159.31 µF 2π × 50 0.0320 = 101.84 µF = ω Cγbc = 0.0320 ⇒ Cγbc = 2π × 50 1 1 = 63.6 mH = = 0.0501 ⇒ Lγca = ω Lγca 2π × 50 × 0.0501 = ω Cγab = 0.0501 ⇒ Cγab =

Total admittances are: Y ab = Ylab +Yγab = 0.0308 + j0.0039  Y bc = Ylbc +Yγbc = 0.0240 + j0  Y ca = Ylca +Yγca = 0.0308 − j0.0039 

Theory of Fundamental Load Compensation

183

Knowing these total admittances, we can find line currents using the following expressions. (b) Source currents compensation Before placing the compensator, the line currents are computed as follows.   I la = I lab − I lca = (1 − α 2 )Ylab − (α − 1)Ylca = 39.6172V ∠0◦ A   I lb = I lbc − I lab = (α 2 − α)Ylbc − (1 − α 2 )Ylab = 32.5573V ∠179.5893◦ A   I lc = I lca − I lbc = (α − 1)Ylca − (α 2 − α)Ylbc = 7.0645V ∠−178.1070◦ A After placing the compensator, the line currents are computed as follows.   I a = I ab − I ca = (1 − α 2 )Y ab − (α − 1)Y ca V = 19.6738∠0◦ A   I b = I bc − I ab = (α 2 − α)Y bc − (1 − α 2 )Y ab V = 19.6738∠−120◦ A   I a = I ca − I bc = (α − 1)Ylca − (α 2 − α)Y bc V = 19.6738∠120◦ A

Thus, we observe that the currents before compensation are unbalanced in magnitude and phase angles, and the currents after compensation are balanced in magnitude and phase angles. (c) Source, compensator, and load powers The load, source, compensator powers, and their power factors are computed as, Sla

= V a (I la )∗ = Pla + jQla = 9.1119 + j0 kVA

Slb

= V b (I lb )∗ = Plb + jQlb = 3.6975 + j6.5116 kVA

Slc

= V c (I lc )∗ = Plc + jQlc = 0.76550 − j1.4332 kVA

Total real power, Pl Total reactive power, Ql

= Pla + Plb + Plc = 13.575 kW = Qla + Qlb + Qlc = 5.0784 kVAr

Power factor in phase-a, p fla

= cos φla = cos(0◦ ) = 1

Power factor in phase-b, p flb

= cos φlb = cos(59.58◦ ) = 0.5062 (lag)

Power factor in phase-c, p flc

= cos φlc = cos(−58.1◦ ) = 0.5283 (lead)

Sa

= V a (I a )∗ = Pa + jQa = 4.5250 + j0.0 kVA

Sb

= V b (I b )∗ = Pb + jQb = 4.5250 + j0.0 kVA

Sc

= V c (I c )∗ = Pc + jQc = 4.5250 + j0.0 kVA

184

Power Quality in Power Distribution Systems: Concepts and Applications

Total real power, P Total reactive power, Q Power factor, p fa

= Pa + Pb + Pc = 13.575 kW = Qa + Qb + Qc = 0 kVAr =

p fb = p fc = cos φa = cos(0◦ ) = 1.0

Compensator powers are given as follows. Sfa Sfb Sfc Total real power, Pf Total reactive power, Q f

= V a (I f a )∗ = V a (I la − I a )∗ = 0 − j 7.9448 kVA = V b (I f b )∗ = V b (I lb − I b )∗ = 0 − j 5.0784 kVA = V c (I f c )∗ = V c (I lc − I c )∗ = 0 + j 7.9448 kVA = Pf a + Pf b + Pf c = 0 kW

= Q f a + Q f b + Q f c = −5.0784 kVAr

From the above results, we observe that, P = Pl = 13.575 kW, Q = 0 kVAr, Pf = 0 kW, and Q f = −Ql = −5.0784 kVAr. This ensures that the load reactive power is fully compensated by the compensator. Also, the compensator does not consume any real power. The source is supplying only the load real power and its reactive power is zero. This makes unity power factor at the source. (d) Equivalent circuit as seen from source and from load side is illustrated in Fig. 230 4.16. The value of equivalent resistance is, Ra = Rb = Rc = 19.6738 = 11.71 Ω or Rab = Rbc = Rca = 3 × 11.69 = 35.07 Ω.

Figure 4.16 Equivalent circuit (a) from source side and (b) from load side

Example 4.7. Consider the following three-phase, three-wire system as shown in Fig. 4.17. The three-phase voltages are balanced sinusoids with rms value of 230 V at 50 Hz. The load impedances are Za = 3 + j 4 Ω, Zb = 5 + j 12 Ω, Zc = 12 − j 5 Ω. Compute the following. (a) The line currents I la , I lb , I lc . (b) The active (P) and reactive (Q) powers of each phase.

Theory of Fundamental Load Compensation

185

bc ca (c) The compensator susceptance ( Bab γ , Bγ , Bγ ), so that the supply sees the load balanced and unity power factor.

(d) For case (3), compute the source, load, compensator active and reactive powers (after compensation).

Figure 4.17 An unbalanced three-phase three-wire star connected load

Solution: Given that Za = 3 + j 4 Ω, Zb = 5 + j 12 Ω, Zc = 12 − j 5 Ω. (a) Line currents I la , I lb and I lc are found by first computing neutral voltage as given below.   1 V aN V bN V cN V nN = 1 + + 1 1 Za Zb Zc Za + Zb + Zc   V aN V bN V cN Za Zb Zc + + = Za Zb + Zb Zc + Zc Za Za Zb Zc 845∠97.82◦ = (24.12∠−99.55◦ ) 252.41∠55.5◦ = 43.79 − j 67.04V = 80.75∠−57.15◦ V Now the line currents are computed as below. I la

=

I lb

=

I lc

=

230∠0◦ − 80.75∠−57.15◦ V aN −V nN = = 39.63∠−33.11◦ A Za 3+ j4 230∠−120◦ − 80.75∠−57.15◦ V bN −V nN = = 15.85∠152.21◦ A Zb 5 + j 12 V cN −V nN 230∠120◦ − 80.75∠−57.15◦ = 23.89∠143.35◦ A = Zc 12 − j 5

(b) Active and reactive load powers For phase-a, Pla Qla

= VaN Ila cos φla = 230 × 39.63 × cos(33.11◦ ) = 7635.9W = VaN Ila sin φla = 230 × 39.63 × sin(33.11◦ ) = 4980VAr

186

Power Quality in Power Distribution Systems: Concepts and Applications

For phase-b, Plb Qlb

= VbN Ilb cos φlb = 230 × 15.85 × cos(−152.21◦ − 120◦ ) = 140.92W

= VbN Ilb sin φlb = 230 × 15.85 × sin(−152.21◦ − 120◦ ) = 3643.2VAr

For phase-c, Plc

= VcN Ilc cos φlc = 230 × 23.89 × cos(−143.35◦ + 120◦ ) = 5046.7W

Qlc

= VcN Ilc sin φlc = 230 × 23.89 × sin(−143.35◦ + 120◦ ) = −2179.3VAr

Total three-phase load powers are given as, = Pla + Plb + Plc = 12823W

Pl Ql

= Qla + Qlb + Qlc = 6443.8VAr

(c) Compensator susceptance First, we convert the star-connected load to a delta load as given below. Zlab

=

Zlbc

=

Zlca

=

Za Zb + Zb Zc + Zc Za = 4 + j 19 = 19.42∠77.11◦ Ω Zc ∆Z = 50.44 + j 2.08 = 50.42∠2.36◦ Ω Za ∆Z = 19.0 − j 4.0 = 19.42∠ − 11.89◦ Ω Zb

The above implies that, Ylab Ylbc Ylca

ab = 1/Zlab = Gab l + j Bl = 0.0106 − j0.050 

bc = 1/Zlbc = Gbc l + j Bl = 0.0198 − j0.0008  ca = 1/Zlca = Gca l + j Bl = 0.0504 + j0.0106 

From the above, the compensator susceptances are computed as follows. Bab γ Bbc γ Bca γ

(Gca − Gbc l √ l ) = 0.0681  3 (Gab − Gca l √ l ) = −0.0222  = −Bbc l + 3 (Gbc − Gab l √ l ) = −0.0053  = −Bca l + 3 = −Bab l +

The respective impedances are, Zγab Zγbc Zγca

= − j 14.69 Ω (capacitance) =

j 45.13 Ω (inductance)

=

j 188.36 Ω (inductance)

187

Theory of Fundamental Load Compensation

(d) After compensation The effective impedances in the delta-connected configuration is, 

Zlab



Zlbc



Zlca

= Zlab || Zγab = 24.97 − j 41.59 = 48.52∠ − 59.01◦ Ω = Zlbc || Zγbc = 21.52 − j 24.98 = 32.97∠49.25◦ Ω

= Zlca || Zγca = 24.97 − j 41.59 = 19.7332∠ − 6.0◦ Ω

Let us convert delta-connected impedances to star-connected impedances. 











Za

=

Zlab × Zlca ◦    = 9.0947 − j 10.55 = 13.93∠ − 49.25 Ω ab Zl + Zlbc + Zlca

Zb

=

Zlbc × Zlab ◦    = 23.15 + j 2.43 = 23.28∠6.06 Ω Zlab + Zlbc + Zlca

Zc

=

Zlca × Zlbc ◦    = 4.8755 + j 8.12 = 9.47∠59.01 Ω Zlab + Zlbc + Zlca

The new voltage between the load and system neutral after compensation is given by, 

V nN

=

V aN V bN V cN 1 +  +  ) = 205.157∠72.8◦ V ( Zb Zc ( Z1 + Z1 + Z1 ) Za a

b

c

Based on the above, the line currents are computed as follows. 

Ia

=

V aN −V nN = 18.584∠0◦ A Za

Ib

=

V bN −V nN = 18.584∠ − 120◦ A Zb

Ic

=

V cN −V nN = 18.584∠120◦ A Zc





Thus, it is seen that after compensation, the source currents are balanced and have unity power factor with respective supply voltages. Source powers after compensation are given as follows. Pa

= Pb = Pc = 230 × 18.584 = 4274.32 W P = 3Pa = 12822.96 W

Qa

= Qb = Qc = 0VAr Q = 0 VAr

188

Power Quality in Power Distribution Systems: Concepts and Applications

The compensator powers are computed as following. ab



bc



ca



ab ∗





bc

ca

2



= V ab (V ab Yγab ) = V abYγab √ = (230 × 3)2 × (− j 0.0068) = − j 10802 VA √ 2 bc ∗ = Vbc Yγ = (230 × 3)2 × ( j 0.0222) = j 3516 VA √ 2 ca ∗ = Vca Yγ = (230 × 3)2 × ( j 0.0053) = j 842 VA = V ab I γ

ab

The total compensator power is Sγ = Sγ +Sγ +Sγ = − j 6444 VA. Thus, we observe that after compensation, the source real power equals to the load power, i.e., Ps = Pl = 12823 W, and source reactive power is zero, i.e., Qs = 0. The compensator supplies the total reactive power of the load, i.e., Qγ = −Ql = −6443.8 VAr, and the compensator real power is zero, i.e., Pγ = 0.

4.7

COMPENSATION OF STAR CONNECTED SYSTEM WITH GROUNDED NEUTRAL

Readers might be curious to know whether it is possible to apply these principles to compensate a three-phase four-wire system where a star-connected load is unbalanced and the neutral is connected to the ground. A three-phase four-wire supply system with an unbalanced star-connected load is shown in Fig. 4.18. Now, the question is whether we can find compensator impedances (Z f a , Z f b , Z f c ) connected between phases and the ground to balance a star-connected unbalanced load. Let the supply

Figure 4.18 A three-phase four-wire system with unbalanced load

voltages be given as follows. vsa (t) = vsb (t) = vsc (t) =

√ √ √

2V sin(ωt) 2V sin(ωt − 120◦ ) ◦

2V sin(ωt + 120 )

(4.129)

189

Theory of Fundamental Load Compensation

Before compensation, the load currents, which are also the source currents, are given as follows. √ ila (t) = 2 Ila sin(ωt − φla ) √ ilb (t) = 2 Ilb sin(ωt − 120◦ − φlb ) (4.130) √ ◦ ilc (t) = 2 Ilc sin(ωt + 120 − φlc ) In the above equations, Rla V Ila =  , φla = cos−1  R2la + Xla2 R2la + Xla2

Rlb V Ilb =  , φlb = cos−1  R2lb + Xlb2 R2lb + Xlb2 Ilc = 

V

R2lc + Xlc2

, φlc = cos−1 

(4.131)

Rlc

R2lc + Xlc2

For Zla = Rla + j Xla , Zlb = Rlb + j Xlb , and Zlc = Rlc + j Xlc . Let the currents after compensation be in phase with their voltages, therefore, √ isa (t) = 2Is sin(ωt) √ isb (t) = 2Is sin(ωt − 120◦ ) (4.132) √ ◦ isc (t) = 2Is sin(ωt + 120 ). In phasor notations, the source voltages and currents are denoted as following. V sa (t) = V ∠0 V sb (t) = V ∠−120◦ V sc (t) = V ∠120

(4.133)



and, I sa (t) = Is ∠0 I sb (t) = Is ∠−120◦

(4.134)

I sc (t) = Is ∠120◦ The total load active power is given as following. Pl







=

Real (V sa I la +V sb I lb +V sc I lc )

=

∗ ∗ ∗ Real(V sa (V sa /Zla ) +V sb (V sb /Zlb ) +V sc (V sc /Zlc ))





= V 2 Real(1/Zla + 1/Zlb + 1/Zlc )



(4.135)

190

Power Quality in Power Distribution Systems: Concepts and Applications

This load power (Pl ) in the above equation must be equal to source power (Ps ), which is given below. Ps = Pl = 3V Is = V 2 Real(1/Zla + 1/Zlb + 1/Zlc ).

(4.136)

The above equation implies that, Is =

V Real(1/Zla + 1/Zlb + 1/Zlc ). 3

(4.137)

From the above, I sa , I sb and I sc can be obtained using (4.134) as following. V sa Real(1/Zla + 1/Zlb + 1/Zlc ) 3 V sb Real(1/Zla + 1/Zlb + 1/Zlc ) I sb = 3 V sc Real(1/Zla + 1/Zlb + 1/Zlc ) I sc = 3 I sa =

(4.138)

The compensator currents, I f a , I f b and I f c are given as following. I fa

= I la − I sa

(−V sa /Z f a ) = V sa /Zla − I sa = V sa /Zla −

V sa Real(1/Zla + 1/Zlb + 1/Zlc ) 3

(4.139)

The above equation implies that, Yf a

= −

1 1 + Real(1/Zla + 1/Zlb + 1/Zlc ) Zla 3

In the above equation, Real(1/Zla + 1/Zlb + 1/Zlc ) = Gla + Glb + Glc Therefore, the expressions for Y f b , Y f b and Y f c can be written as following. Yf a Yf b Yf c

1 = −Yla + (Gla + Glb + Glc ) 3 1 = −Ylb + (Glb + Glb + Glc ) 3 1 = −Ylc + (Gla + Glb + Glc ) 3

(4.140)

191

Theory of Fundamental Load Compensation

Or, Yf a

=

Yf b

=

Yf c

=

1 (−2 Gla + Glb + Glc ) − j Bla 3 1 G f b + j B f b = (Gla − 2 Glb + Glc ) − j Blb 3 1 G f c + j B f c = (Gla + Glb − 2 Glc ) − j Blc 3

(4.141)

1 Rla (−Xla ) = 2 +j 2 = Gla + j Bla Zla Rla + Xla2 Rla + Xla2

(4.142)

Rfa (−X f a ) 1 = 2 +j 2 = Gfa + jBfa Zfa R f a + X f2a R f a + X f2a

(4.143)

Gfa + jBfa =

In the above equations, Yla

=

and Yf a

=

Similarly, the above relations can be given for phase-b and phase-c load and filter admittances. Based on (4.141), the compensator conductances and suseptances are given as follows. Gfa

=

Gfb

=

Gfc

=

1 (−2 Gla + Glb + Glc ) 3 1 (Gla − 2 Glb + Glc ) 3 1 (Gla + Glb − 2 Glc ) 3

(4.144)

and, B f a = −Bla B f b = −Blb B f c = −Blc

(4.145)

Once G f a , B f a , G f b , B f b , G f c , and B f c are known from (4.144)–(4.145), the resistive and reactive part of the compensator network can be computed using following expressions. Zfa

=

Rfa + j Xfa =

Gfa (−B f a ) +j 2 G2f a + B2f a G f a + B2f a

Zfb

=

Rfb + j Xfb =

Gfb 2 G f b + B2f b

+j

Zfc

=

Rfc + j Xfc =

Gfc 2 G f c + B2f c

+j

(−B f b ) G2f b + B2f b

(−B f c ) G2f c + B2f c

(4.146)

192

Power Quality in Power Distribution Systems: Concepts and Applications

Comparing (4.144)–(4.145) with (4.127), we find that the compensator susceptance is negative of load susceptance in both the schemes of compensation. However, in the former scheme, the compensator network is of purely reactive parameters, and hence the scheme is practically realizable, as illustrated in Example 4.5. But in power system network where neutral is grounded, the compensator impedances have both real and reactive parts as given by (4.146). This indicates that it is not possible to compensate the unbalanced load by using purely reactive network alone. The second aspect which is important is that the resistive parts of the compensator impedances, i.e., R f a , R f b , and R f c are not all positive. One or two of them are of negative sign, as clear from (4.144). This can be further illustrated from the losses point of view. Since total average load active power comes from source, implying that there are no losses in the compensator. That is, I 2f a R f a + I 2f b R f b + I 2f c R f c = 0

(4.147)

Since the terms, I 2f a , I 2f a , and I 2f a are positive, indicating that one or two terms out of three, i.e., R f a , R f b , and R f c must be of negative signs. This means that there are negative resistances, which can not be realized in practice by using passive components. Negative resistances values appear in (4.146) to replenish the losses due to positive resistance. Even if all resistances are positive, making compensator impedances realizable is not a practical scheme, as irreversible losses occur. These losses have to be supplied from some active source which is not possible with passive components. Therefore, one of the possible schemes in this case, is to set the conductance part of compensator admittance to zero, i.e., G f a = G f b = G f c = 0. This is given below. Yf a Yf b

= Gfa + jBfa = jBfa

Yf c

= Gfc + jBfc = jBfc

= Gfb + jBfb = jBfb

Using (4.145), B f a = −Bla , B f b = −Blb , B f c = −Blc , therefore the overall admittance looking into the source terminal is given by, Ysa = Yla +Y f a Ysb = Ylb +Y f b

= Gla + j Bla − j Bla = Gla

Ysc = Ylc +Y f c

= Glc + j Blc − j Blc = Glc

= Glb + j Blb − j Blb = Glb

(4.148)

The impedance seen by the source is given as follows. Zsa Zsb Zsc

R2 + Xla2 1 = la Gla Rla R2lb + Xlb2 1 = = Rsb = Glb Rlb R2lc + Xlc2 1 = Rsc = = Glc Rlc = Rsa =

(4.149)

Theory of Fundamental Load Compensation

193

Thus, the source sees a three-phase resistive load as shown in Fig. 4.19. The source

Figure 4.19 A three-phase four-wire system compensated system

currents for this scheme of compensation are given below. I sa

V sa Rla V sa R  la = V sa 2 = 2 Rsa Rla + Xla R2la + Xla2 R2la + Xla2   Rla  = Ila cos φla ∠V sa = Ila ∠V sa   2 Rla + Xla2 =

(4.150)

Similarly, source currents for the other two phases can be given by the following expressions.   R lb  = Ilb cos φlb ∠V sb I sb = Ilb ∠V sb   (4.151) R2lb + Xlb2   R lc  = Ilc cos φlc ∠V sc I sc = Ilc ∠V sc   (4.152) 2 2 Rlc + Xlc Obviously, the source currents for this compensated system are in phase with their voltages to give a unity power factor. As seen from (4.150)–(4.152), magnitudes of the source currents are directly proportional to the power factor in respective phases, and therefore their magnitudes are reduced. Consequently, the current in the neutral wire is also reduced. Since the currents are not balanced in magnitude due to different values of resistances seen from the source, the objective of the compensation is not fully achieved. Thus, this scheme provides partial compensation. That is why in the case of the grounded system, to balance the load, active filtering is used by employing active power filters. These devices are called custom power devices [16]– [18]. The one which is used for current compensation and connected with the load in the shunt is called shunt active power filter, which is also known as distribution static compensator (DSTATCOM). These types of load compensation schemes and the devices will be discussed in the next chapter.

194

Power Quality in Power Distribution Systems: Concepts and Applications

Example 4.8. Consider the following three-phase, four-wire system as shown in Fig. 4.18. The three-phase voltages are balanced sinusoids with rms value of 230 V at 50 Hz. The load impedances are Za = 5 + j 12 Ω, Zb = 3 + j 4 Ω, Zc = 12 − j 13 Ω. Answer the following. (a) The line currents, I la , I lb , I lc , I nN . (b) The total active (P) and reactive (Q) load power of the load supplied by the source. (c) Compute compensated source currents (I sa , I sb , I sc ), compensator currents (I f a , I f b , I f c ) and compensator impedances (Z f a , Z f b , Z f c ), so that the supply sees a balanced load with unity power factor. (d) For case (c), compute total active and reactive source powers and compensator powers. Observe that Ps = Pl , Qs = 0, Pf = 0 and Q f = Ql . (e) Explain why this scheme is not practically realizable. (f) Discuss the scheme which is realizable using purely reactive components in the compensator. Find out these values of compensator impedances and corresponding network parameters. (g) For case (f), calculate source currents and neutral current and compare these values with those obtained in (c). Solution: Given that Za = 5 + j 12 Ω, Zb = 3 + j 4 Ω, Zc = 12 − j 13 Ω. (a) Line currents, I la , I lb , I lc are calculated as below. I la

=

I lb

=

I lc

=

V sa 230∠0◦ = 17.69∠−67.38◦ A = Zla 5 + j 12 V sb 230∠−120◦ = 46.00∠−173.13◦ A = Zlb 3+ j4 V sc 230∠120◦ = 13.00∠167.29◦ A = Zlc 12 − j 13

The neural current, without compensation, is given as follows. I nN = I la + I lb + I lc = 54.93∠−159.79◦ A (b) Active (P) and reactive (Q) powers of the load supplied by the source are given as follows. Sla



= Pla + j Qla = V a I la = 230∠0◦ × (17.69∠−67.38◦ )∗ = 1565.1 + j 3756.2 VA

Slb



= Plb + j Qlb = V b I lb = 230∠−120◦ × (46.00∠−173.13◦ )∗ = 6348 + j 8464 VA

Theory of Fundamental Load Compensation

Slc

= =

195



Plc + j Qlc = V c I lc = 230∠120◦ × (13.00∠167.29◦ )∗ 2028.1 − j 2197.1 VA

Total real power, Pl = Pla + Plb + Plc = 9941.2 W Total reactive power, Ql = Qla + Qlb + Qlc = 10023.1 VAr (c) With Compensator The supply should see the load as a balanced and unity power factor. Therefore the source currents are computed using (4.138) and are given below. I sa

= = = = =

I sb

= = = =

I sc

= = = =

V sa Real(1/Zla + 1/Zlb + 1/Zlc ) 3 1 1 1 230∠0◦ Real( + + ) 3 5 + j 12 3 + j 4 12 − j 13 230∠0◦ Real(0.1879 − j 0.1895) 3 230∠0◦ (0.1879) 3 14.4075∠0◦ A V sb Real(1/Zla + 1/Zlb + 1/Zlc ) 3 1 1 1 230∠−120◦ Real( + + ) 3 5 + j 12 3 + j 4 12 − j 13 230∠−120◦ (0.1879) 3 14.4075∠−120◦ A V sc Real(1/Zla + 1/Zlb + 1/Zlc ) 3 1 1 1 230∠120◦ Real( + + ) 3 5 + j 12 3 + j 4 12 − j 13 230∠120◦ (0.1879) 3 14.4075∠120◦ A

As seen from the above calculation, the source currents are balanced and in phase with their respective phase voltages. The source neutral current is zero, as source currents are balanced. Once the source currents are known, the compensator currents are computed as follows. I fa

=

I fb

=

I fc

=

I la − I sa = 17.69∠−67.38◦ − 14.41∠0◦ = 18.01∠−114.96◦ A

I lb − I sb = 46.00∠−173.13◦ − 14.41∠−120◦ = 39.09∠169.72◦ A I lc − I sc = 13.00∠167.29◦ − 14.41∠120◦ = 11.07∠−119.66◦ A

196

Power Quality in Power Distribution Systems: Concepts and Applications

The corresponding impedances are now computed as below. Zfa

= −

230∠0◦ V sa =− = 5.38 − j 11.57 Ω 18.01∠−114.96◦ I fa

Zfb

= −

230∠−120◦ V sb =− = −1.98 − j 5.54 Ω 39.09∠169.72◦ I fb

Zfc

= −

V sc 230∠120◦ =− = 10.49 + j 17.93 Ω 11.07∠−119.66◦ I fc

Therefore, the compensator susceptances are given as the following. Yf a

=

Yf b

=

Yf c

=

1 = 0.033 + j 0.071  Zfa 1 = −0.057 + j 0.16  Zfb 1 = 0.024 − j0.041  Zfc

(d) Active (P) and reactive (Q) powers of the source after compensation Ssa



= Psa + jQsa = V a I sa = 230∠0◦ × (14.4075∠0◦ )∗ = 3313.7 VA

Ssb



= Psb + jQsb = V b I sb = 230∠−120◦ × (14.4075∠−120◦ )∗ = 3313.7 VA

Ssc

= Psc + jQsc = V c (I sc )∗ = 230∠120◦ × (14.4075∠120◦ )∗ = 3313.7 VA

Total real power, Ps = Psa + Psb + Psc = 9941.2 W Total reactive power, Qs = Qsa + Qsb + Qsc = 0 VAr The active and reactive powers of the compensator are computed as follows. Sfa Sfb



= Pf a + jQ f a = V a I f a = 230∠0◦ × (18.01∠−114.96◦ )∗ = −1748.6 + j 3756.2 VA ∗

= Pf b + jQ f b = V b I f b = 230∠−120◦ × (39.09∠169.72◦ )∗ = 3034.3 + j 8464.0 VA

Sfc



= Pf c + jQ f c = V c I f c = 230∠120◦ × (11.07∠−119.66◦ )∗ = −1285.6 − j 2197.1 VA

Total real power, Pf = Pf a + Pf b + Pf c ≈ 0 W Total reactive power, Q f = Q f a + Q f b + Q f c = 10023.1 VAr Thus, it is seen that Ps = Pl , Qs = 0, Pf = 0 and Q f = Ql .

Theory of Fundamental Load Compensation

197

(e) The compensator impedances are Z f a = 5.38 − j 11.57 Ω, Z f b = −1.98 − j 5.54 Ω, Z f c = 10.49 + j 17.93 Ω. It can be observed that the compensator impedances have real parts which correspond to the positive and negative resistances. Due to the presence of these negative resistances, it is not possible to realize this scheme. (f) A practical realizable compensator As the compensator impedances have real parts which correspond to positive or negative resistances and are therefore not realizable. Thus to overcome this problem, the real parts of the compensator susceptances are set to zero. That is, Zfa

=

Zfb

=

Zfc

=

1 = − j 14.08 Ω j Imag(Y f a ) 1 = − j 6.25 Ω j Imag(Y f b ) 1 = j 24.07 Ω j Imag(Y f c )

These are realized by placing the capacitors in phases a and b, and the inductor in phase c. These values are given as follows. Cfa

=

Cfb

=

Lfc

=

1 = 226.02 µF 2 π × 50 × 14.08 1 = 509.3 µF 2 π × 50 × 6.25 24.07 = 76.6 mH 2 π × 50

(g) The source currents and neutral current for this scheme of compensation are given below. I sa I sb I sc I nN

Rla = 6.80∠0 A R2la + Xla2 Rlb = V sb Glb = V sb 2 = 27.60∠ − 120◦ A Rlb + Xlb2 Rlc = V sc Glb = V sc 2 = 8.82∠120◦ A Rlc + Xlc2 = V sa Gla = V sa

= I sa + I sb + I sc = 19.86∠ − 125.03◦ A

In (g), rms values of source currents are 6.8, 27.6, and 8.82 A in phase-a, b, and c, respectively. Considering the feeder of identical resistance, r, in each line, the 2 + I 2 + I 2 + I 2 ) = 1280.21 r W. For part (c), the feeder loss will be Pf loss = r × (Isa sc nN sb source currents are balanced with rms value of 14.41 A. This gives a feeder loss of 3 Is2 r = 622.2 r W. Thus, the feeder losses in (c) are approximately half that of the

198

Power Quality in Power Distribution Systems: Concepts and Applications

case in (g). This is because the source currents in (g) are partially compensated and hence remain unbalanced. The feeder losses in the uncompensated system are quite 2 + I 2 + I 2 + I 2 ) = 5615.1 r W, which is approximately high, i.e., Pf loss = r × (Ila nN lb lc nine times to those in (c). This emphasizes the fact that the power system should be operated with balanced voltage and current quantities for its efficient and economic operation.

4.8

SUMMARY

In this chapter, the theory of load compensation of passive linear load is explained. It is inferred that a compensator can function independently, either as a load compensator or as a voltage regulator. Various stages of compensator currents are realized using, sampling and averaging techniques, and expressions for compensator susceptances are developed in terms of load currents at specified phase voltages. It is found that for delta-connected load or three-phase ungrounded system, it is always possible to design a compensator comprised of purely reactive elements such as inductance and capacitance to draw source currents which are balanced sinusoids at the fundamental frequency and in phase with the utility voltages. It is proved that this is not true for three-phase grounded system as it involves real power requirements from the compensator phases. This problem is addressed using active power filters, which will be discussed in the next chapter.

4.9

PROBLEMS

P 4.1 Describe various kinds of loads in the power system by giving their specific examples. What are their impacts on power system performance and characteristics? P 4.2 Explain the term voltage regulation at the load bus in the power system. Show that the load bus voltage is a function of the load current drawn by the load. Give examples where the voltage regulation could be positive, negative, and zero. P 4.3 Write down some of the practical aspects and characteristics of the compensator, used to correct the power factor, voltage and current unbalances in the system. P 4.4 Explain how can locally generated reactive power by a compensator regulate the bus voltage and improve the power factor at the load bus. P 4.5 How can the reactive power by a compensator be generated at the load bus? Which type of power system elements generates or absorbs the reactive power? Give some practical examples for applications of these elements. P 4.6 Define the short circuit capacity and short circuit current of a bus in a power system. Develop an approximate expression for load bus voltage in terms of

Theory of Fundamental Load Compensation

199

system voltage, load reactive power, and short circuit capacity. Explain the nature of variation of bus voltage as load reactive power varies. P 4.7 Mention characteristics of the ideal voltage regulator and derive the expression for bus voltage (V ) in terms of the supply voltage (E), compensator gain (Kγ), knee point (Vk ), load reactive power (Ql ), and short circuit capacity (Ssc ) of the bus. P 4.8 Comment upon the neutral voltage and neutral current in three-phase, threewire, and three-phase, four-wire electrical systems. Explain in three-phase, four-wire system, why is grounding necessary. P 4.9 What is the relationship between zero sequence voltage (current) and neutral voltage (current) quantities? P 4.10 Consider a three-phase supply system feeding 11 kV (L-L), 50 Hz, load with the real and reactive power demand of 40 MW and 50 MVAr, respectively. The supply bus has short-circuit level of 300 MVA and feeder with a Xs /Rs ratio of 6. Compute the following. (a) Find the load bus voltage (V ) and the voltage drop (∆V ) in the supply feeder. Thus, determine system voltage (E). (b) It is required to maintain the load bus voltage to be the same as the supply bus voltage, i.e., VLL = 11 kV. Calculate the rating of the compensator. P 4.11 Consider a three-phase supply system at 11 kV line-to-line voltage with a short circuit level of 500 MVA and a Xs /Rs ratio of 7, supplying a star-connected inductive load with a real and reactive power demand of 50 MW and 75 MVAr, respectively. (a) It is required to maintain the load bus voltage to be the same as the supply bus voltage i.e., VLL = 11 kV. Calculate the rating of the compensator, compensator reactance, and the nature of its reactive element. (b) What should be the load bus voltage and compensator current if it is required to maintain the unity power factor at the load bus? P 4.12 A three-phase distribution feeder is designed to deliver a three-phase induction motor load of 20 MW at 11 kV (L-L), 50 Hz, and a power factor of 0.707 lagging. Now, 1 kA of extra load at a power factor of 0.5 lagging for each phase is connected to the load bus. (a) What are the total requirement of real power (P) in MW and reactive (Q) in MVAr at the load bus? (b) Now, this total load with the above real (P) and reactive (Q) power demands in (a), is supplied by a three-phase supply at 11 kV, 50 Hz. The supply bus has short-circuit level of 300 MVA and a feeder with a Xs/Rs ratio of 5. Compute the following.

200

Power Quality in Power Distribution Systems: Concepts and Applications

(i) The voltage at the load bus. (ii) The reactive power supplied by the compensator connected to the load bus to maintain the same load bus voltage (i.e., 11 kV (L-L)) and the value of compensator element per phase. (iii) For unity power factor operation at the load bus, what is the rating of the compensator and voltage at the load bus? P 4.13 For a delta-connected load shown in Fig. 4.15, load admittances are: Ylab = ab bc bc bc ca ca ca Gab l + j Bl , Yl = Gl + j Bl and Yl = Gl + j Bl . Derive relationship for bc ab ca compensator susceptances, (Bγ , Bγ , Bγ ), such that the line currents are balanced and have unity power factor with their voltages. Assume the source voltages to be balanced and sinusoidal. Consider the load parameters as given below. Zlab = Y1ab = 5 + j 12, Zlbc = Y1bc = 3 + j 4, Zlca = Y1ca = 9 − j 12 Ω. l

l

l

(a) Compute line currents before the placing compensator. (b) Compute susceptances of the compensator in order to achieve unity power factor at the source. (c) Determine the source currents after placing the compensator. (d) Determine the source and compensator active and reactive powers. Corelate these powers with the load powers. P 4.14 A three-phase system with 400 V rms (line to line) voltage, 50 Hz, is supplying a resistive load connected between phases a and b with R = 10 Ω. A purely reactive compensator is connected between the phases. Compute the line currents, active and reactive powers, power factors of each phase, and as well bc ca as three-phase. Now compute the compensator susceptances, (Bab γ , Bγ , Bγ ) so that the supply has a balanced load with unity power factor. Co-relate the source, load, and compensator active and reactive powers. Find the rating of the compensator. Also, draw the circuit seen from the source and load side. P 4.15 Consider the following three-phase, three-wire system as shown in Fig. 4.20. The three-phase voltages are balanced sinusoids with rms value of 230 V. The values of load impedances are given as, Za = 3 + j 4 Ω, Zb = 5 + j 12 Ω and Zc = 20 + j 15 Ω.

Figure 4.20 Related to problem P 4.15

Theory of Fundamental Load Compensation

201

Compute the following. (a) Line currents. (b) Total active (P) and reactive (Q) powers of the load. ab ca (c) Compensator susceptances (Bbc γ , Bγ , Bγ ), so that the supply sees the load balanced and unity power factor.

(d) Find source currents after compensation. (e) For case (c), compute the total source, compensator active and reactive powers (after compensation). Show that compensator supplies the reactive power and source supplies active power of the load. (f) Draw the circuits as seen from the load and from the source. ab ca P 4.16 Obtain and verify the compensator susceptances (Bbc γ , Bγ , Bγ ) of the above problem using Sampling Method. ab ca P 4.17 Obtain and verify the compensator susceptances (Bbc γ , Bγ , Bγ ) of above the problem using Averaging Method.

P 4.18 Sampling and Averaging methods give you real time values of compensator susceptances to achieve a unity power factor on the source side. How would you practically realize the variable susceptances in real time? P 4.19 A three-phase system supplies three-phase ungrounded star connected load with Za = 10 + j 15 Ω, Zb = 15 + j 20 Ω, and Zc = ∞ Ω, (phase-c open) by balanced voltages with rms value of 230 V at 50 Hz to get a unity power factor at the supply. bc ca (a) Compute these susceptances (Bab γ , Bγ , and Bγ ) and express them in terms of physical parameters, i.e., inductance or capacitance and their values.

(b) Determine the source currents after placing the compensator. (c) Determine the source, compensator, and load active and reactive powers. (d) Draw the circuits as seen from the load and from the source side. P 4.20 Consider the following three-phase, four-wire system. The three-phase voltages are balanced sinusoids with rms value of 230 V at 50 Hz. The load impedances are Za = 12 + j 5 Ω, Zb = 6 − j 8 Ω, Zc = 3 + j 4 Ω. Compute the following. (a) Determine the total active (P) and reactive (Q) load power of the load supplied by the source. (b) Compute the compensator currents (I f a , I f b , I f c ) and impedances (Z f a , Z f b , Z f c ), so that the supply current see the load balanced and unity power factor. (c) Compute the source currents before and after compensation.

202

Power Quality in Power Distribution Systems: Concepts and Applications

(d) With above compensator parameters, show that, Ps = Pl , Qs = 0, Pf = 0 and Q f = Ol . (e) Comment upon the realization of calculated compensator impedances. P 4.21 In problem P 4.20, realize the compensator with pure reactive elements and determine the following. (a) Compute the values of the elements of the reactances. (b) Determine the source and neutral currents and comment upon their values.

REFERENCES 1. A. Ghosh and G. Ledwich, Power Quality Enhancement Using Custom Power Devices, ser. Power Electronics and Power Systems. Springer US, 2012. 2. R. W. Erickson and D. Maksimovi´c, Power and Harmonics in Nonsinusoidal Systems. Boston, MA: Springer US, 2001, pp. 589–607. 3. M. H. J. Bollen, Understanding Power Quality Problems: Voltage Sags and Interruptions. Wiley-IEEE Press, 1999. 4. Bhim Singh, Ambrish Chandra, and Kamal Al-Haddad, Power Quality Problems and Mitigation Techniques. Wiley, 2015. 5. Roger C. Dugan, Mark F. McGranaghan, Surya Santoso, and H. Wayne Beaty, Electrical Power System Quality. Tata McGraw-Hill, 2008. 6. M. A. Masoum and E. F. Fuchs, Chapter 1 – Introduction to Power Quality, M. A. Masoum and E. F. Fuchs, Eds. Academic Press, 2015. 7. M. Grady, Understanding Power System Harmonics. Department of Electrical and Computer Engineering, University of Texas, Austin, 2012. 8. A. Baggini, Ed., Handbook of Power Quality. Wiley Sons, England, 2008. 9. B. W. Kennedy, Power Quality Primer. Springer, 2000. 10. C. Sankaran, Power Quality. CRC Press, 2002. 11. G. T. Heydt, Electric Power Quality. Stars in a Circle Publications, 1991. 12. T. J. E. Miller, Reactive Power Control in Electric Systems. Wiley, 1982. 13. L. Gyugyi, “Reactive power generation and control by thyristor circuits,” IEEE Transactions on Industry Applications, vol. IA-15, no. 5, pp. 521–532, 1979. 14. R. Otto, T. Putman, and L. Gyugyi, “Principles and applications of static, thyristor-controlled shunt compensators,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-97, no. 5, pp. 1935–1945, 1978.

Theory of Fundamental Load Compensation

203

15. N. Hingorani and L. Gyugyi, Understanding FACTS: Concepts and Technology of Flexible AC Transmission Systems, 2000. Wiley-IEEE Press, 1999. 16. M. El-Habrouk, M. K. Darwish, and P. Mehta, “Active power filters: a review,” Electric Power Applications, IEE Proceedings, vol. 147, no. 5, pp. 403–413, 2000. 17. H. Akagi, Y. Kanazawa, and A. Nabae, “Instantaneous reactive power compensators comprising switching devices without energy storage components,” IEEE Transactions on Industry Applications, vol. IA-20, no. 3, pp. 625–630, 1984. 18. H. Akagi, “Trends in active power line conditioners,” Power Electronics, IEEE Transactions on, vol. 9, no. 3, pp. 263–268, 1994.

Theories for Load 5 Control Compensation 5.1

INTRODUCTION

Originally, passive filters were evolved for high voltage dc transmission systems in order to reduce harmonic voltages and currents in ac power networks to acceptable levels. Later on, these were also used for balancing an unbalanced load. However, passive filters have certain disadvantages, such as increased size, lack of adaptability, inability to re-route active power in order to balance the load, etc., which are overcome by active power devices (filters). Active power filters have been proposed by eminent researchers [1]–[5]. Recent progress in voltage and current rating and switching speed of semiconductor devices such as IGBTs, GTO thyristors, has spurred interest in studying active power filters with a focus on practical applications. The sophisticated PWM inverter technology along with developed control methods [6]–[8] have made it possible to put active power filters into commercial installations. In the last couple of decades, researchers have developed many efficient theories for shunt compensation to control active shunt devices. The main purpose of a shunt active power compensator is to cancel the effects of poor load power factor, harmonic currents, and unbalance in the load. The dc component in load current can also be compensated by choosing a suitable compensator configuration. As a result of this compensation, source currents are balanced and sinusoidal with desired power factor. In this chapter, we shall discuss various control theories of shunt compensation, starting from sampling and averaging techniques proposed in 1970s [9], [10], to the later ones, i.e., instantaneous reactive power theory [6], and theory of instantaneous symmetrical components [11]–[13]. In the previous chapter, we studied the methods of load compensation at the fundamental frequency with a linear passive load consisting of resistance, inductance, and capacitance. These methods can eliminate only the fundamental reactive power and unbalance in the steady state for three-phase three-wire systems by employing reactive passive LC filters and thyristor controlled devices. However, when harmonics are present in the system, these methods fail to provide correct compensation. To correct load with unbalance and harmonics, instantaneous load compensation methods are used. The important theories in this context are the Instantaneous Reactive Power Theory, often known as pq theory, the Instantaneous Symmetrical Component Theory for load compensation and the Synchronous Reference Frame based theory (also known as dq theory) [3], [6], [7], [11], [14]–[21]. The first two theories will be discussed in this chapter due to their wide applications in load compensation. Their merits, demerits and applications are explored in detail.

DOI: 10.1201/9781032617305-5

204

205

Control Theories for Load Compensation

5.2

INSTANTANEOUS REACTIVE POWER THEORY

To begin with pq theory, we shall first recall the αβ 0 transformation, which was discussed in Chapter 2. For three-phase system shown in Fig. 5.1, the αβ 0 transformations for voltages and currents are given below.

Figure 5.1 A three-phase four-wire system



v0



  vα  =   vβ   i0   iα  =   iβ







√1  2

2 1 3 0 

√1  2

2 1 3 0

 

(5.1)

 

(5.2)

√1 2 − 12 √ 3 2

√1 v 2   a  vb  − 12  √   vc − 23

√1 2 − 12 √ 3 2

√1 i 2   a 1  i  − 2   b √ ic − 23

The instantaneous active, p(t) and reactive, q(t) powers were defined in Chapter 3 through equations (3.14)–(3.15), respectively. For the sake of completeness, these are given below. p3φ (t) =

va ia + vb ib + vc ic

=

vα iα + vβ vβ + v0 i0

=

pα + pβ + p0

=

pαβ + p0

where, pαβ = pα + pβ = vα iα + vβ iβ and p0 = v0 i0 .

(5.3)

206

Power Quality in Power Distribution Systems: Concepts and Applications

In the instantaneous reactive power theory, as discussed in Chapter 3, the instantaneous reactive power, q(t) was defined as, q(t) = qαβ

= vα × iβ + vβ × iα

= vα iβ − vβ iα 1 = − √ [vbc ia + vca ib + vab ic ] 3

(5.4)

Therefore, powers po , pαβ and qαβ can be expressed in matrix form as given below.      v0 p0 0 0 i0  pαβ  =  0 vα vβ  iα  (5.5) iβ qαβ 0 −vβ vα

From the above equation, the currents, i0 , iα and iβ are computed as given below.    −1   v0 p0 0 0 i0 iα  =  0 vα vβ   pαβ  iβ 0 −vβ vα qαβ    2 2 0 0 vα + vβ p0 1  0 (5.6) = v0 vα −v0 vβ   pαβ  v0 (v2α + v2β ) qαβ 0 v0 vβ v0 vα

In the above equation,

i0



=

=

p0 (v2α + v2β ) v0 (vα 2 + vβ 2 )

=

v 0 i0 p0 = = i0 v0 v0

(−vβ ) vα pαβ + 2 q vα + vβ 2 αβ v2α + v2β

= iα p + iαq iβ

=

(5.7)

v2α

(5.8)

vβ vα pαβ + 2 q 2 2 αβ v + vβ α + vβ

= iβ p + iβ q where, i0 = zero sequence instantaneous current iα p = α-phase instantaneous active current = iβ p = β -phase instantaneous active current =

(5.9)

vα v2α +v2β vβ v2α +v2β

iαq = α-phase instantaneous reactive current = iβ q = β -phase instantaneous reactive current =

pαβ pαβ

vβ − v2 +v 2 qαβ α β vα q . 2 2 vα +vβ αβ

Control Theories for Load Compensation

207

Using the above definitions of various components of currents, the three-phase instantaneous power can be expressed as, p3φ

=

v0 i0 + vα iα + vβ iβ

=

v0 i0 + vα (iα p + iαq ) + vβ (iβ p + iβ q )   −vβ vα v0 i0 + vα 2 pαβ + 2 qαβ vα + v2β vα + v2β   vβ vα +vβ 2 pαβ + 2 qαβ vα + v2β vα + v2β

=

=

(5.10)

v0 i0 + vα iα p + vα iαq + vβ iβ p + vβ iβ q

= v0 i0 + (pα p + pαq ) + (pβ p + pβ q ) = v0 i0 + (pα p + pβ p )

(5.11)

In the above equation, pαq + pβ q = vα iαq + vβ iβ q = 0 If referred to compensator (or filter), (5.6) can be written as,    2   0 0 vα + v2β pf0 if0 1  0 i f α  = v0 vα −v0 vβ   p f αβ  2 + v2 ) v (v 0 α β ifβ q f αβ 0 v0 vβ v0 vα

(5.12)

(5.13)

Since the compensator does not supply any real power, therefore, p f 3φ = p f 0 + p f αβ = 0

(5.14)

The instantaneous zero sequence power exchanges between the load and the compensator and compensator reactive power must be equal to load reactive power. Therefore, we have p f o = plo = plo + plo = vo ilo p f αβ = plαβ − plo

(5.15)

p f o + p f αβ = plo + plαβ

(5.18)

p f o + p f αβ = 0

(5.19)

q f αβ = qlαβ = vα ilβ − vβ ilα

(5.16) (5.17)

Since the overall real power from the compensator is equal to zero (the time varying active power can still be there), therefore the following should be satisfied.

and,

208

Power Quality in Power Distribution Systems: Concepts and Applications

The power flow description is shown in Fig. 5.2. In the figure, ∆p = plo + Ploss , is additional real power to be supplied from the source to maintain dc link voltage constant. The term Ploss represents losses in the inverter. The zero sequence current should be circulated through the compensator, therefore, i f o = ilo

(5.20)

Using the conditions of compensator powers as given in above, the α and β compo-

Figure 5.2 Power flow description of three-phase four-wire compensated system

nents of compensator currents can be given as follows. ifα

= = =

−vβ vα p f αβ + 2 q f αβ v2α + v2β vα + v2β v2α

1 [vα (−vo ilo ) − vβ (vα ilβ − vβ ilα )] + v2β

1 [−vα vo ilo − vβ vα ilβ + v2β ilα )] v2α + v2β

(5.21)

Similarly, ifβ

=

v2α

1 [vβ ( p f αβ ) + vα ( q f αβ )] + v2β

=

1 [vβ (−vo ilo ) + vα (vα ilβ − vβ ilα )] v2α + v2β

=

1 [−vo vβ ilo − vα vβ ilα + v2α ilβ ] v2α + v2β

(5.22)

Control Theories for Load Compensation

209

The above equations are derived based on the assumption that in general, vo = 0. If vo = 0, then if0

=

ilo

ifα

=

1 [v2 ilα − vα vβ ilβ ] v2α + v2β β

ifβ

=

v2α

(5.23)

1 [−vα vβ ilα + v2α ilβ ] + v2β

Once the compensator currents, i f o , i f α and i f β are known, they are transformed back to the abc frame in order to implement in real time. This transformation is given below.   1   ∗  √ 1 0  ifa 2 √  ifo  2 3   √1 − 1 i∗f b  =  (5.24) 2 2√  i f α 3  12 ∗ 3 1 i ifc f β √ −2 − 2 2

These reference currents are shown in Fig. 5.3. Once reference compensator currents are known, these are tracked using a voltage source inverter (VSI). The other details of the scheme are given as follows.

Figure 5.3 A three-phase four-wire compensated system with ideal compensator

The power components and various components of currents are related as follows.     vα iα pα = pβ vβ iβ   vα (iα p + iαq ) = vβ (iβ p + iβ q )     vα iα p v i = + α αq vβ iβ p vβ iβ q     p pα p + αq (5.25) = pβ p pβ q

210

Power Quality in Power Distribution Systems: Concepts and Applications

The following quantities are defined. α - axis instantaneous active power = pα p = vα iα p α - axis instantaneous reactive power = pαq = vα iαq β - axis instantaneous active power = pβ p = vβ iβ p β - axis instantaneous reactive power = pβ q = vβ iβ q It is seen that, pα p + pβ p

= vα iα p + vβ iβ p = vα =

v2α + v2β v2α + v2β

vβ vα pαβ + vβ 2 pαβ v2α + v2β vα + v2β

pαβ = pαβ

(5.26)

and, pαq + pβ q = vα iαq + vβ iβ q = vα

−vβ vα qαβ + vβ q =0 vα 2 + vβ 2 vα 2 + vβ 2 αβ

(5.27)

Thus, it can be observed that the sum of pα p and pβ p is equal to total instantaneous active power pαβ , and the sum of pαq and pβ q is equal to zero. Therefore, p3φ

=

pαβ + po

=

pα + pβ + po

=

pα p + pβ p + po

(5.28)

For a practical compensator, the switching and ohmic losses should be considered. These losses should be met from the source in order to maintain the dc link voltage constant. Let these losses be denoted by Ploss , and the average power that must be supplied to the compensator be ∆p, as illustrated in Fig. 5.2, then ∆p is given as following. ∆p = plo + Ploss

(5.29)

Now, the compensator powers can be expressed as, pf0

=

p f αβ q f αβ

=

pl0 plαβ − ∆p

(5.30)

= qlαβ

With the above compensator powers, the reference compensator currents are computed as follows.    2   0 0 vα + v2β plo ifo 1  0 i f α  = vo vα −v0 vβ   p˜lαβ − ∆p (5.31) 2 + v2 ) v (v 0 α β ifβ qlαβ 0 v0 vβ v0 vα

211

Control Theories for Load Compensation

Knowing these currents, we can obtain compensator currents in the abc frame using the following equation.   1 ∗    √ 1 0  ifa 2 √  ifo  2 3 1 1   √ − i∗f b  = (5.32) 2 2√  i f α 3  12 ∗ 3 1 i ifc f β √ − − 2

2

2

Figure 5.4 A neutral clamped voltage source inverter

These reference currents are realized using a voltage source inverter (VSI). One of the common VSI topologies, which is used to track the reference currents, is illustrated in Fig. 5.4. This VSI topology is known as neutral clamped inverter. The topology has two dc storage capacitors (Cdc1 , Cdc2 ) with voltages, Vdc1 and Vdc2 , respectively. Together these capacitors form the dc link with voltage, Vdc = Vdc1 +Vdc2 for the inverter. It consists of six switches (S1 to S6 ) with anti-parallel diodes (D1 to D6 ) to provide a freewheeling path for the respective phase currents, as shown in the figure. The switches shown in the figure are realized using Insulated Gate Bipolar Transistors (IGBTs) or Metal Oxide Field Effect Transistors (MOSFETs) depending on their voltage, current, switching frequency and other important parameters. The voltage source inverter is connected to the point of common coupling (PCC) through interface inductors (L f − R f ) in each phase [22]. A schematic diagram of a threephase four-wire compensated system is shown in Fig. 5.5. As explained above, the reference compensator currents, i.e., i∗f a , i∗f b , i∗f c are generated using an active filter controller. The actual compensator currents, i.e., i f a , i f b , i f c are measured using current transducers. These currents are compared with respective phase reference currents within the defined hysteresis band. This is indicated by dynamic hysteresis current control block, which generates switching signals for the voltage source inverter. While realizing the compensator using a voltage source inverter, there are switching and other losses in the inverter circuit. Therefore, a voltage regulator is

212

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 5.5 Control scheme for three-phase four-wire compensated system

used to maintain the dc link voltage to its reference value. This voltage regulator is basically a proportional integral (PI) controller, which generates Ploss term to account for the losses in the inverter circuit. Example 5.1. Consider a three-phase four-wire system as shown in Fig. 5.3. The supply voltages are balanced with rms value of 230 V at 50 Hz. The load impedances are: Zla = 5 + j 8 Ω, Zlb = j 4 Ω and Zlc = − j 12 Ω. Using instantaneous reactive power theory, calculate the following. (a) αβ 0 components of three-phase supply voltages (vsα , vsβ , vs0 ,) and load currents (ilα , ilβ , il0 ). (b) The various components of instantaneous and average powers i.e., pα , pβ , p0 , qαβ . (c) Verify these powers on the basis of real power values, i.e., Pabc = Pα + Pβ + P0 = Pαβ 0 and reactive power, Qabc = Qαβ . (d) Reference compensator currents in αβ 0 frame i.e., (i f α , i f β , i f 0 ) and abc frame i.e., (i f a , i f b , i f c ) so that source supplies balanced currents from the source with unity power factor relationship with their voltages. (e) Check whether the source currents are balanced and in phase with their voltages. (f) Check whether ps (t) = Pl , qs (t) = 0, p f (t) = 0 and q f (t) = ql .

213

Control Theories for Load Compensation

Solution: (a) The three-phase voltages and currents are expressed as follows. √ vsa (t) = 2V sin(ωt) √ vsb (t) = 2V sin(ωt − 120◦ ) √ vsc (t) = 2V sin(ωt + 120◦ ) ila (t) = ilb (t) = ilc (t) =

√ 2Ila sin(ωt − φa ) √ 2Ilb sin(ωt − 120◦ − φb ) √ 2Ilc sin(ωt + 120◦ − φc )

In phasor form, for V = 230 V, the voltages and currents are given as follows. V sa

=

230∠0◦ V

V sb

=

V sc

=

230∠ − 120◦ V

I la

=

I lb

=

I lc

=

230∠120◦ V

V sa = 24.38∠ − 58◦ A 5+ j8 V sb = 57.5∠ − 210◦ A j4 V sc = 19.16∠210◦ A − j 12

Based on the above current phasors, the rms values of currents (Ila , Ilb , Ilc ) and their phase angles (φa , φb , φc ) are given as following. Ila = 24.38 A, Ilb = 57.5 A, Ilc = 19.16 A, φa = −58◦ , φb = 90◦ , φc = −90◦ . Since the voltages and currents have only fundamental components, the calculation of αβ 0 components can be made using phasors. Based on the phasor values, we can write time expressions. These are given in the following.     2 V sb V sc 3 = V sα = V sa − − V ∠0◦ = 281.7 ∠0◦ V 3 2 2 2   √ 2 3 3 (V sb −V sc ) = V ∠−90◦ = 281.7 ∠−90◦ V V sβ = 3 2 2 1 V s0 = √ (V sa +V sb +V sc ) = 0 V 3

214

Power Quality in Power Distribution Systems: Concepts and Applications

Similarly, I lα I lβ I l0

   2 I lb I lc − = 45.04∠−33.27◦ A = I la − 3 2 2  √ 2 3 = (I lb − I lc ) = 35.85 ∠130.89◦ A 3 2 1 = √ (I la + I lb + I lc ) = 30.85 ∠−178.38◦ A 3

Based on the above equations, time expressions of αβ 0 components of threephase supply voltages are given as follows.   √ vsb vsc  2 vsa − = 281.7 2 sin ωt vsα = − 3 2 2  √ √ 2 3 vsβ = (5.33) (vsb − vsc ) = 281.7 2 sin(ωt − 90◦ ) 3 2 1 vs0 = √ (vsa + vsb + vsc ) = 0 3 Similarly, the time expressions of αβ 0 components of three-phase load currents are given as follows.    √ ilb ilc 2 ila − − = 45.04 2 sin(ωt − 33.27◦ ) ilα = 3 2 2  √ √ 2 3 (ilb − ilc ) = 35.85 2 sin(ωt + 130.89◦ ) ilβ = 3 2 √ 1 il0 = √ (ila + ilb + ilc ) = 30.85 2 sin(ωt − 178.38◦ ) 3 Therefore, we have the following voltages and currents in αβ 0 phasor form, V sα

= 281.7∠0◦ V

V sβ

= 281.7∠ − 90◦ V

V s0

= 0V

and the currents are as follows. I lα I lβ I l0

= 45.04∠ − 33.27◦ A = 35.85∠130.89◦ A

= 30.85∠−178.38◦ A

215

Control Theories for Load Compensation

(b) Power components plα , plβ , pl0 , qlαβ plα

√ √ = vsα ilα = 281.7 2 sin ωt 45.04 2 sin(ωt − 33.27◦ ) = 12.68 {cos 33.27◦ − cos(2ω t − 33.27◦ )}

=

10.61 − 12.68 cos(2ω t − 33.27◦ ) kW

Therefore, the average value of plα = Plα = 10.61 kW Oscillating value of plα = plα = −12.68 cos(2ω t − 33.27◦ ) kW. √ √ plβ = vsβ ilβ = 281.7 2 sin(ωt − 90◦ ) 35.85 2 sin(ωt + 130.89◦ ) = 10.09 {cos 220.89◦ − cos(2ω t + 40.89◦ )} =

−7.64 − 10.09 cos(2ω t + 40.89◦ ) kW

Average value of plβ = Plβ = −7.63 kW Oscillating value of plβ = plβ = −10.09 cos(2ω t + 40.89◦ ) kW.

Therefore, the total active power in αβ frame i.e., plαβ is given as, plαβ

=

plα + plβ

=

2.97 + 18.24 cos(2ω t + 178.86◦ )kW

(5.34)

Average value of plαβ = Plαβ = 2.97 kW. Oscillating value of plαβ = plαβ = 18.24 cos(2ω t + 178.86◦ ) kW. The zero sequence power is zero as vs0 = 0. Therefore, pl0

= vs0 il0 = 0

The instantaneous reactive power in αβ 0 frame is given by, qlαβ

vsα ilβ − vsβ ilα √ √ = 281.7 2 sin(ωt) 35.85 2 sin(ωt + 130.89◦ ) √ √ −281.7 2 sin(ωt − 90◦ ) 45.04 2 sin(ωt − 33.27◦ ) = 10.09 {cos 130.89◦ − cos(2ω t + 130.89◦ )} −12.69 {cos 56.73◦ − cos(2ω t − 123.27◦ )} =

=

−6.61 − 10.09 cos(2ω t + 130.89◦ ) − 6.96

+12.69 cos(2ω t − 123.27◦ )

= −13.57 − 18.24 cos(2ω t + 88.89◦ ) kVAr From above expressions, the average value of these terms is given below. P3φ = Plα + Plβ + P0 Qlαβ

=

2.97 kW

=

−13.57 kVAr

216

Power Quality in Power Distribution Systems: Concepts and Applications

(c) The three-phase average real and reactive powers of the load are given by, P3φ = Vsa Ila cos φa +Vsb Ilb cos φb +Vsc Ilc cos φc = 2.97 + 0 + 0 = 2.97 kW Q3φ = −(Vsa Ila sin φa +Vsb Ilb sin φb +Vsc Ilc sin φc ) = 4.75 + 13.23 − 4.41 = 13.57 kVAr Thus, we observe that, P3φ

= Pabc = Pαβ 0 = 2.97 kW

Q3φ

= Qabc = −Qαβ = 13.57 kVAr

(d) The required filter powers, p f 0 , p f αβ and q f αβ in terms of load powers are given below. pfo

=

p f αβ

=

q f αβ

= qlαβ = −13.57 − 18.24 cos(2ω t + 88.89◦ ) kVAr

plo = 0 plαβ = 18.24 cos(2ω t + 178.86◦ ) kW

Now the filter currents can be found by using (5.13) and substituting the above filter powers. These are expressed below. if0 ifα ifβ

= il0 = 43.63 sin(ωt − 178.38◦ ) A −vβ vα p f αβ + 2 q f αβ = 57.59 sin(ωt − 37.36◦ ) A = v2α + v2β vα + v2β vβ vα = p f αβ + 2 q f αβ = 56.52 sin(ωt + 126.93◦ ) A 2 2 vα + vβ vα + v2β

Substituting these values in equation (5.32), we get filter currents in abc frame. These are expressed below. √ ifa = 2 × 22.39 sin(ωt − 67.38◦ ) A √ ifb = 2 × 57.66 sin(ωt + 145.72◦ ) A √ ifc = 2 × 19.65 sin(ωt − 137.33◦ ) A (e) Therefore, the source currents isa , isb , isc after compensation are given below. √ isa = ila − i f a = 2 × 4.3 sin ωt A √ isb = ilb − i f b = 2 × 4.3 sin(ωt − 120◦ ) A √ isc = ilc − i f c = 2 × 4.3 sin(ωt + 120◦ ) A It is seen that the above source currents are balanced and in phase with the respective voltages.

217

Control Theories for Load Compensation

(f) From the above values of voltage and currents of source, load and compensator, the powers are given below. ps qs pf qf

5.2.1

vsa isa + vsb isb + vsc isc = 2.97 kW = Pl 1 = − √ (vsbc isa + vsca isb + vsab isc ) = 0 kVAr 3 = vsa i f a + vsb i f b + vsc i f c = 0 kW 1 = − √ (vsbc i f a + vsca i f b + vsab i f c ) 3 = −13.57 − 18.24 cos(2ω t + 88.89◦ ) kVAr = ql =

(5.35)

STATE SPACE MODELING OF THE COMPENSATOR

There are various VSI topologies that can be used to realize DSTATCOM [23]. The most commonly used is neutral clamped inverter topology as shown in Fig. 5.4. Since this is a three-phase four-wire system each phase can be considered independently. Therefore, to analyze the above circuit, only one phase is considered. When switch S1 is closed and S4 is open, as shown in Fig. 5.6(a), then Kirchhoff’s Voltage Law (KVL) is given in the following equation.

Figure 5.6 Equivalent circuit (a) S1 (ON), S4 (OFF) and (b) S1 (OFF), S4 (ON)

Lf

di f a + R f i f a + vsa −Vdc1 dt

=

0

(5.36)

From the above equation, di f a dt

=



Rf vsa Vdc1 ifa − + Lf Lf Lf

(5.37)

Similarly, when S1 is open and S4 is closed as shown in Fig. 5.6(b), di f a dt

=



Rf vsa Vdc2 ifa − − Lf Lf Lf

(5.38)

The above two equations can be combined into one equation by using switching signals Sa = S1 , Sa = S4 , (S1 and S4 are complementary binary signals for switches

218

Power Quality in Power Distribution Systems: Concepts and Applications

in leg-a of the inverter), as given below. di f a dt

= −

Rf Vdc1 Vdc2 vsa i f a + Sa − Sa − Lf Lf Lf Lf

(5.39)

Similarly, for phases b and c, the first order derivative of filter currents can be written as follows. di f b dt di f c dt

Rf Vdc1 Vdc2 vsb i f b + Sb − Sb − Lf Lf Lf Lf Rf Vdc1 Vdc2 vsc = − i f c + Sc − Sc − Lf Lf Lf Lf

(5.40)

= −

(5.41)

where, Sa = 0 and Sa = 1 implies that the top switch is open and bottom switch is closed and, Sa = 1 and Sa = 0 implies that the top switch is closed and bottom switch is open. The two logic signals Sa and Sa are complementary to each other. This logic also holds for the other two phases. The inverter currents i1 and i2 , as shown in Fig. 5.4, can be expressed in terms of filter currents and switching signals. These are given below. i1 i2

= Sa i f a + Sb i f b + Sc i f c Sa i f a + Sb i f b + Sc i f c

=

(5.42)

The relationship between dc capacitor voltages Vdc1 , Vdc2 and inverter currents i1 and i2 are given as below. dVdc1 dt dVdc2 Cdc2 dt Cdc1

= −i1 (5.43)

= i2

Considering Cdc1 = Cdc2 = Cdc and substituting i1 and i2 from (5.42), the above equations can be written as, dVdc1 dt dVdc2 dt

Sb Sc Sa ifa − ifb − ifc Cdc Cdc Cdc Sa Sb Sc ifa + ifb + ifc Cdc Cdc Cdc

= −

(5.44)

=

(5.45)

The equations (5.39), (5.40), (5.41), (5.44) and (5.45) can be represented in state space form as given below.  R  Sa Sa    − 1   − L ff 0 0 − L L 0 0 f f Lf ifa ifa   Rf S S   b − Lf 0 − Lbf   i f b   0 − L1f 0  i   0 Lf  vsa  d   fb     R     S S f 1 c c  ifc  = 0 − L f  vsb 0 0 − Lf − Lf   i f c  +  0 Lf dt V    Sa  Vdc1   0 Sb Sc dc1 0 0  vsc 0  − Cdc − Cdc − Cdc 0 Vdc2

Sa Cdc

Sb Cdc

Sc Cdc

0

0

Vdc2

0

0

0

(5.46)

219

Control Theories for Load Compensation

The above equation is of the form, x˙ = A x + B u

(5.47)

where, x is a state vector, A is system matrix, B is input matrix and u is input vector. This state space equation can be solved using MATLAB to implement the compensator for the simulation study. 5.2.2

SWITCHING CONTROL OF THE VSI

In (5.46), the switching signals Sa , Sa , Sb , Sb , Sc , and Sc are generated using a hysteresis band current control. This is described as follows. The upper and lower bands of the reference filter current (say phase-a) are formed using hysteresis h, i.e., i∗f a + h and i∗f a − h. The following logic is used to generate switching signals. If i f a ≥ (i∗f a + h)

Sa = 0, and Sa = 1 else if i f a ≤ (i∗f a − h)

Sa = 1, and Sa = 0 else if (i∗f a − h) < i f a < (i∗f a + h)

Retain the current status of the switches

end The first-order derivative of state variables can be easily solved using “c2d” (continuous to discrete) command in MATLAB, as given below. [Ad Bd ] = c2d(A, B, Td )

(5.48)

The value of the state vector is updated using the following equation. x[(k + 1)Td ] = Ad x[kTd ] + Bd u[kTd ]

(5.49)

where x(k + 1) refers the value of the state vector at (k + 1)th sample or at the instant t = (k + 1)Td for Td as the time duration between two consecutive samples. The solution of state equation given by (5.47) is given as follows [24]. x(t) = e

A(t−t0 )

x(t0 ) +

t

eA(t−τ) B u(τ)dτ

to

where to represents initial time and t represents current time.

(5.50)

220

Power Quality in Power Distribution Systems: Concepts and Applications

Writing the above equation for a small time interval, kTd ≤ t ≤ (k +1)Td with to = kTd and t = (k + 1)Td , x[(k + 1)Td ]

x[kTd ] +

(k+1)T  d

= eATd x[kTd ] +

(k+1)T  d

ATd

= e

eA {(k+1)Td −τ} B u[kTd ] dτ

kTd

kTd

  eA {(k+1)Td −τ} B dτ u[kTd ]

(5.51)

Comparing (5.49) and (5.51), the discrete matrices Ad and Bd computed by “c2d” MATLAB function can be written as follows. Ad Bd

= eA Td =

(k+1)T  d

eA {(k+1)Td −τ} B dτ

(5.52)

kTd

5.2.3

GENERATION OF PLOSS TO MAINTAIN DC CAPACITOR VOLTAGE

The next step is to determine Ploss in order to maintain the dc link voltage close to its reference value. In compensation, the average voltage variation of dc link is an indicator of Ploss in the inverter. If losses are more than the power supplied by the source through inverter, the dc link voltage, i.e., Vdc = Vdc1 +Vdc2 , will decrease toward zero and vice-versa. For proper operation of the compensator, we need to maintain the dc capacitor voltage to two times of the reference value of each capacitor voltage, i.e., Vdc1 + Vdc2 = Vdc = 2Vdcre f . The value of Vdcre f is chosen well above (approximately 1.5 times) the peak value of the system voltage [25]. Thus, we have to replenish losses in the inverter and sustain the dc capacitor voltage to 2Vdcre f with each capacitor voltage at Vdcre f . This is achieved by using proportional integral (PI) controller described below [7]. Let us define an error signal as follows. eV dc = 2Vdcre f − (Vdc1 +Vdc2 ) = 2Vdcre f −Vdc Then, the term Ploss is computed as follows. Ploss

= K p eV dc + Ki

 Td 0

eV dc dt

This dc link voltage control loop need not be very fast and may be updated once in a voltage cycle, preferably at the positive zero crossings of phase-a voltage and generate Ploss term at these points. The above controller can be implemented using the digital domain as follows. Ploss (k) =

k

K p eVdc (k) + Ki ∑ eVdc ( j) Td . j=0

(5.53)

221

Control Theories for Load Compensation

In the above equation, k represents the kth sample of error, eVdc . For k = 1, the above equation can be written as, 1

Ploss (1) = K p eVdc (1) + Ki ∑ eVdc ( j) Td j=0

=

K p eVdc (1) + Ki [eVdc (0) + eVdc (1)] Td

(5.54)

Similarly for k = 2, we can write, Ploss (2)

=

  K p eVdc (1) + Ki eVdc (0) + eVdc (1) + eVdc (2) Td

Replacing Ki [eVdc (0) + eVdc (1)] from (5.54), we get,   Ki eVdc (0) + eVdc (1) Td = Ploss (1) − K p eVdc (1)

(5.55)

(5.56)

Substituting above value in (5.55), we obtain the following. Ploss (2)

= =

K p eVdc (2) + Ploss (1) − K p eVdc (1) + Ki eVdc (2) Td   Ploss (1) + K p eVdc (2) − eVdc (1) + Ki eVdc (2) Td

(5.57)

In general, for kth sample of Ploss , Ploss (k)

=

  Ploss (k − 1) + Kp eVdc (k) − eVdc (k − 1) + Ki eVdc (k) Td (5.58)

The above algorithm can be used to implement PI controller to generate Ploss . The control action of the dc link voltage control loop can be updated once at every positive zero crossing of phase-a voltage. 5.2.4

COMPUTATION OF LOAD AVERAGE POWER

In reference current expressions, the average load power (Plavg ) is required to be computed. Although a low pass filter can be used to find this, its dynamic response is slow, which may not be suitable for effective compensation. The dynamic performance of computation of Plavg plays a significant role in compensation. For this reason, a moving average algorithm can be used to compute average power, which is described below. Plavg =

1 T

 T 0

(va ila + vb ilb + vc ilc ) dt

(5.59)

The above equation can be written as an integration operation from t1 − T to t1 as given in the following. Plavg

=

1 T

 t1

t1 −T

(va ila + vb ilb + vc ilc ) dt

(5.60)

This is known as moving average filter (MAF). Any change in variables instantly reflects within the settling time of one cycle. In the above equation, T represents the time period of fundamental component of voltage or current. The time t1 is any arbitrary time instant.

222

5.3

Power Quality in Power Distribution Systems: Concepts and Applications

SOME MISCONCEPTIONS IN INSTANTANEOUS REACTIVE POWER THEORY

The instantaneous reactive power theory has evolved from Fortescue, Clarke and Park Transformations of voltage and current specified in phases-a, b and c coordinates [26]. In general, for three-phase, four-wire system,      √1 √1 √1  va vo 2 2 2      2  1 − 1 − 1   vb  vα  = (5.61)   3 √2 √2    3 vβ vc − 23 0 2 Similarly, for currents, the αβ 0 components are given as follows.      √1 √1 √1  i io 2 2 2   a   2  1 − 1 − 1   ib  iα  =   3 √2 √2    3 3 iβ ic 0 − 2 2

(5.62)

√ For balanced system v0 = (va + vb + vc )/ 3 = 0. For three-phase three-wire system, ia + ib + ic = 0, which implies that i0 = 0. Using these details, the above transformations in (5.61) and (5.62) result in the following.      va   1 1  vα 2 1 −2 −2   vb  √ √ = (5.63)   3 3 3 0 vβ − 2 2 vc   iα iβ

  2 1 − 12 √ = 3 3 0 2

   ia  − 12   ib  √   − 23 ic

From equation (5.63), and using va + vb + vc = 0, we get the following.      2 2 vb + vc vb vc  va − = va − − vα = 3 2 3 2 2     va vb 2 vb va − − − − = 3 2 2 2  3 = va 2

(5.64)

(5.65)

223

Control Theories for Load Compensation

  √ √   √ √ 2 3 3 2 3 3 vb − vc = vb − (−va − vb ) = 3 2 2 3 2 2  √ √ √  2 3 3 3 vb + va + vb = 3 2 2 2 √ 1 = √ v a + 2 vb 2



Writing (5.65) and (5.66) in matrix form we get,       3 0 vα v =  12 √  a vβ √ 2 vb

(5.66)

(5.67)

2

Similary, using ia + ib + ic = 0 the following can be written.       3 0 iα ia 2   √ = 1 iβ √ 2 ib 2

(5.68)

According to the pq theory, the abc components of voltages and currents are transformed to the α and β coordinates and the instantaneous powers p and q of the load can be expressed as follows. p =

vα iα + vβ iβ

(5.69)

q =

vα iβ − vβ iα

(5.70)

The above equations representing instantaneous active and reactive powers can be expressed in the matrix form as following [6].      vα vβ iα p (5.71) = q −vβ vα iβ Therefore, from the above equation (5.71), the α β components of currents can be expressed as follows.   iα iβ The matrix,



vα −vβ

vβ vα 

−1

vα −vβ

=



vα −vβ

vβ vα

−1   p q

is given as following. vβ vα

−1

 1 vα = 2 2 vα + vβ vβ

−vβ vα



(5.72)

224

Power Quality in Power Distribution Systems: Concepts and Applications

From the above equations, iα

=



=

vβ vα p− 2 q 2 + vβ vα + v2β vβ vα p+ 2 q 2 vα + v2β vα + v2β v2α

(5.73) (5.74)

Which can further be written as, iα

= iα p + iαq

(5.75)



= iβ p + iβ q

(5.76)

In the above equation, iα p iαq iβ p iβ q

vα p v2α + v2β vβ = − 2 q vα + v2β vβ = p 2 vα + v2β vα = q 2 vα + v2β =

(5.77) (5.78) (5.79) (5.80)

The instantaneous active and reactive components of currents in supplying line can be calculated from the α and β components of the current as given in the following.  −1       2  3 0 0 iα p + iαq i ia α 2 3 √  =  1 = iβ ib √ − √16 √12 iβ p + iβ q 2 2        2 2 0 0 iα p iαq 3 3 = + 1 1 1 1 i βp − √6 √2 − √6 √2 iβ q    2 iap 3 = 1 ibp − √6

0 √1 2

  iα p iβ p

(5.81)

  iαq iβ q

(5.82)

and,    2 iaq 3 = ibq − √16

0 √1 2

The active and reactive components of the line currents must be consistent to the basic definitions. However, these components of currents have little in common with the reactive power of the load as defined in [26]. This is shown in the following illustration.

225

Control Theories for Load Compensation

Example 5.2. Consider a resistive load connected as shown in Fig. √ 5.7. It is supplied from a three-phase balanced sinusoidal voltage source with va = 2V sin ωt, with V = 230 V. Express the voltage and currents for primary and secondary side of the transformer. Express the active and reactive component of the currents, powers and discuss about them.

Figure 5.7 An unbalanced resistive load supplied by three-phase delta-star connected transformer

Solution: With the above given values, the primary side phase voltages with respect to virtual ground could be expressed as the following. √ √ vA = 2V sin ωt = 230 2 sin ωt √ √ vB = 2V sin(ωt − 120◦ ) = 230 2 sin(ωt − 120◦ ) (5.83) √ √ ◦ ◦ vC = 2V sin(ωt + 120 ) = 230 2 sin(ωt + 120 ) In phasor form, VA

=

230∠0◦ V

VB

=

230∠−120◦ V

VC

=

230∠120◦ V

Therefore, the primary side line-to-line voltages are expressed as follows. √ √ 2 3V sin(ωt + 30◦ ) vAB = √ √ vBC = 2 3V sin(ωt − 90◦ ) √ √ vCA = 2 3V sin(ωt + 150◦ ) In phasor form, V AB

= 398.37∠30◦ V

V BC

= 398.37∠−90◦ V

V CA

= 398.37∠150◦ V

(5.84)

226

Power Quality in Power Distribution Systems: Concepts and Applications

These voltages are transformed into the secondaries and are expressed below. √ √ √ va = 2 3V sin(ωt + 30◦ ) = 398.37 2 sin(ωt + 30◦ ) √ √ √ vb = 2 3V sin(ωt − 90◦ ) = 398.37 2 sin(ωt − 90◦ ) (5.85) √ √ √ ◦ ◦ vc = 2 3V sin(ωt + 150 ) = 398.37 2 sin(ωt + 150 ) In phasor form, Va

= 398.37∠30◦ V

Vb

= 398.37∠−90◦ V

Vc

= 398.37∠150◦ V

Therefore, the currents on the secondary side are given below. √ √ 2 3V sin(ωt + 30) ia = R ib = 0 ic

(5.86)

= 0

Taking V = 230 V and R = 4 Ω, the currents on the secondary side of the transformer are given as follows. √ √ va 2 3V = sin(ωt + 30◦ ) ia = R √ 4 √ = 99.59 2 sin(ωt + 30◦ ) = 2I sin(ωt + 30◦ ) ib

= 0

ic

= 0

(5.87)

In phasor form, the above can be expressed as, Ia

= 99.59∠30◦ A

Ib

= 0A

Ic

= 0A

(5.88)

This phase-a current, ia in the secondary side of the transformer, is transformed to the primary of the delta-connected winding, therefore the currents on the primary side of the transformer are given as follows. √ iA = 2I sin(ωt + 30◦ ) √ iB = −iA = − 2I sin(ωt + 30◦ ) (5.89) iC = 0

227

Control Theories for Load Compensation

The above can be written in phasor form as given below. IA

= 99.59∠30◦ = I∠30◦ A

IB

= −I A = 1 ∠−180◦ × 99.59∠30◦ = 99.59∠ − 150◦ = I∠ − 150◦ A

IC

(5.90)

= 0A

After, knowing the voltages and currents of the primary side of the transformer, their α and β components are expressed as.       3 0 vα v =  12 √  A (5.91) vβ √ 2 vB 2

Substituting vA and vB from (5.83) in the above equation, we get the following.   √     √   3 0 vα 2V sin ωt 3V sin ωt 2   √ √ √ = = (5.92) vβ − 3V cos ωt 2V sin(ωt − 120◦ ) √1 2 2

and,

   3 iα =  12 iβ √ 2



  0 i √  A 2 iB



=



 √ 0  √2 I sin(ωt + 30◦ ) √ ◦ √1 2 − 2 I sin(ωt + 30 ) 2 √  3 I sin(ωt + 30◦ ) (5.93) −I sin(ωt + 30◦ )

= 

3 2

Based on the above transformation matrix, the active and reactive powers are computed as, vα iα + vβ iβ √ √ √ = 3V sin ωt 3I sin(ωt + 30◦ ) − 3V cos ωt[−I sin(ωt + 30◦ )] √  √ 1 3 = 2 3V I sin(ωt + 30◦ ) sin ωt + cos ωt 2 2 √   3V I 2 sin2 (ωt + 30◦ ) = √ p(t) = 3V I [1 − cos 2(ωt + 30◦ )] (5.94) p(t)

q(t)

q(t)

=

vα iβ − vβ iα √ √ √ = 3V sin ωt [−I sin(ωt + 30◦ )] − (− 3V cos ωt) 3I sin(ωt + 30◦ )   √ √ 3 1 ◦ = −2 3V I sin(ωt + 30 ) sin ωt − cos ωt 2 2 √ = −2 3V I sin(ωt + 30◦ )(− cos(ωt + 30◦ )) √ 3V I sin 2(ωt + 30◦ ) (5.95) = =

228

Power Quality in Power Distribution Systems: Concepts and Applications

Based on the above values of p and q powers, the α and β components of active and reactive components are given below.     i +i iα = α p αq iβ iβ p + iβ q where, iα p

=

vα p v2α + v2β

√ 3V sin ωt √ √ p ( 3V sin ωt)2 + (− 3V cos ωt)2 √ 1 = √ sin ωt 3V I [1 − cos 2(ωt + 30◦ )] 3V = I sin ωt [1 − cos 2(ωt + 30◦ )]

=

(5.96)

Similarly, iβ p

=

v2α

vβ p + v2β

√ − 3V cos ωt √ √ p = ( 3V I sin ωt)2 + (− 3V cos ωt)2 √ − 3V cos ωt √ = 3V I [1 − cos 2(ωt + 30◦ )] 3V 2 = −I cos ωt [1 − cos 2(ωt + 30◦ )] iαq

−vβ q + v2β √ −(− 3V cos ωt) √ = 3V I sin 2(ωt + 30◦ ) 3V 2 = I cos ωt sin 2(ωt + 30◦ ) =

iβ q

(5.97)

v2α

= = =

vα q v2α + v2β √ 3V sin ωt √ 3V I sin 2(ωt + 30◦ ) 3V 2 I sin ωt sin 2(ωt + 30◦ )

(5.98)

(5.99)

Thus, knowing iα p , iαq , iβ p and iβ q , we can determine active and reactive components of currents on the source side as given below.      −1 iα p iAp = C iβ p iBp

229

Control Theories for Load Compensation

where, 

[C]−1 = 

−1  2 0 3 √  =  −1 √ 2 6

3 2

√1 2

0 √1 2

 

Using the above equation, we can find out the active and reactive components of the current, as given below.      2 ◦ )} 0 {1 I sin ωt − cos 2(ωt + 30 iAp  =  3 −1 √ √1 −I cos ωt {1 − cos 2(ωt + 30◦ )} iBp 6

2

From the above, iAp

= = = =

iBp

= =

= = = = = iAq

= =



2 I sin ωt {1 − cos 2(ωt + 30◦ )} 3  2 I {2 sin ωt − 2 sin ωt cos 2(ωt + 30◦ )} 32 I √ {2 sin ωt − sin(3ωt + 60◦ ) − sin(−ωt − 60◦ )} 6 I √ {2 sin ωt + sin(ωt + 60◦ ) − sin(3ωt + 60◦ )} 6 1 1 − √ iα p + √ iβ p 6 2 1 − √ I sin ωt {1 − cos 2(ωt + 30◦ )} 6 1 + √ (−I cos ωt) {1 − cos 2(ωt + 30◦ )} 2   √ 2I 1 3 ◦ − √ {1 − cos 2(ωt + 30 )} sin ωt + cos ωt 2 2 6 2I − √ {1 − cos 2(ωt + 30◦ )} sin(ωt + 60◦ ) 6 2I − √ {sin(ωt + 60◦ ) − sin(ωt + 60◦ ) cos 2(ωt + 30◦ )} 6   2I 1 1 − √ sin(ωt + 60◦ ) + sin(ωt) − sin(3ωt + 120◦ ) 2 2 6 I − √ {sin ωt + 2 sin(ωt + 60◦ ) − sin(3ωt + 120◦ )} 6   2 2 I {2 sin 2(ωt + 30◦ ) cos ωt} iαq = 3 32 I √ {sin(ωt + 60◦ ) + sin(3ωt + 60◦ )} 6

230

Power Quality in Power Distribution Systems: Concepts and Applications

iBq

1 1 = − √ iαq + √ iβ q 6 2 1 1 = − √ I cos ωt sin 2(ωt + 30◦ ) + √ I sin ωt sin 2(ωt + 30◦ ) 6 2   √ I 3 1 = − √ 2 sin 2(ωt + 30◦ ) − cos ωt + sin ωt 2 2 6 I = √ 2 sin 2(ωt + 30◦ ) sin(ωt − 30◦ ) 6 I = √ (cos(ωt + 90◦ ) − cos(3ωt + 30◦ )) 6 I = √ {− sin ωt − cos(3ωt + 30◦ )} 6

Thus, we have, iAp iBp iAq iBq

I √ {2 sin ωt + sin(ωt + 60◦ ) − sin(3ωt + 60◦ )} 6 I = − √ {sin(ωt) + 2 sin(ωt + 60◦ ) − sin(3ωt + 120◦ )} 6 I = √ {sin(ωt + 60◦ ) + sin(3ωt + 60◦ )} 6 I = √ {− sin ωt − cos(3ωt + 30◦ )} . 6 =

(5.100) (5.101) (5.102) (5.103)

From the above equations, the terms instantaneous active current and instantaneous reactive current given in the pq theory do not have commonality with the notion of active and reactive currents used in electrical engineering. Also, the reactive current iq occurs in the supply lines of the load in spite of the absence of any reactive element in the load. Furthermore, the nature of the load is linear, and harmonics are absent. From (5.100) to (5.103), it is observed that the net currents in phase-A and B are obtained by adding the active and reactive terms, i.e., iA = iAp + iAq and iB = iBp + iBq , as given in (5.89). These currents do not have any harmonics, however the resolutions of active and reactive components of the currents (iAp , iAq , iBp , iBq ), based on pq theory have harmonics. For example, in the above discussion, I iAp = √ {2 sin ωt + sin(ωt + 60◦ ) − sin(3ωt + 60◦ )} 6

(5.104)

is the active current component in the phase-A, and it contains the third order harmonic. This contradicts the basic notion of the active current that was introduced to electrical engineering by Fryze [26]. Thus, it seems that there is a misconception of electrical phenomenon in three-phase circuits with balanced sinusoidal voltages for linear loads that do not have harmonics. Moreover, the active and reactive currents, iAp , iAq , iBp , iBq which result from the pq theory, are not the current that should remain

231

Control Theories for Load Compensation

in the supply lines after the load is compensated to unity power factor as defined by Fryze [26]. Also, it is evident that the instantaneous reactive power q(t) as defined by the pq theory, does not really identify the power properties of load instantaneously. For example, for the above discussion, the active and reactive powers are given as follows. √ p(t) = 3V I {1 − cos 2(ωt + 30◦ )} √ 3V I sin 2(ωt + 30◦ ) q(t) = Hence, the following points are noted. 1. The active components of currents (iAp , iBp , iCp ) and reactive components of currents, (iAq , iBq , iCq ) contain third harmonic, which is not possible for a linear load as discussed above. 2. The instantaneous reactive power q(t) defined by pq theory does not identify with the power properties of the load instantaneously. Both powers p(t) and q(t) are time varying quantities, so that a pair of their values at any single point of time does not identify with the power properties of the load. The possibility of instantaneous identification of active and reactive power p(t) and q(t) does not mean that power properties of load are identified instantaneously. For example,   √ p(t) = 2 3V I ◦ ◦ at (ωt + 30 ) = 90 , q(t) = 0 The above implies that as if it is a resistive load.   p(t) = 0 Similarly at (ωt + 30◦ ) = 0◦ , q(t) = 0 which implies as there is no load. When (ωt + 30◦ ) = 105◦ ,

  √ p(t) = 3V I (1 + 32 ) √ q(t) = − 3V I( 12 )



implies that it is capacitive load. ◦



Similarly when (ωt + 30 ) = 75 ,



  √ p(t) = 3V I (1 + 32 ) √ q(t) = 3V I( 21 )

implies as if the load is inductive. We, therefore, conclude that power properties cannot be identified without monitoring p(t) and q(t) powers over the entire cycle period. For example, in the above case, the instantaneous reactive power q(t) has occurred, not because of the load reactive elements, but because of load currents unbalance. This unbalance nature of load can

232

Power Quality in Power Distribution Systems: Concepts and Applications

not be identified by instantaneous reactive power q(t). Therefore, pq theory does not offer an advantage with respect to the time interval needed to identify the nature of load and its property over the power theories based on the time domain or frequency domain approaches that require the system to be monitored over one time period. Power computations The secondary side powers are given as following. ∗

Sa Pa

= Pa + jQa = V a I a = 398.37∠30◦ 99.59∠ − 30◦ = 39675 VA

Sb

= Pb + jQb = V b I b = 398.37∠ − 90◦ × 0 = 0 VA

Pb Sc Pc

= 39675 W, Qa = 0 VAr ∗

= 0 W, Qb = 0 VAr ∗

= Pc + jQc = V c I c = 398.37∠150◦ × 0 = 0 VA = 0 W, Qc = 0 VAr

The total active and reactive powers on the secondary side are given as following. P

= Pa + Pb + Pc = 39675 W

Q

= Qa + Qb + Qc = 0 VAr

Svect

= Sarith = P = 39675 VA

p fvect

=

p farith = P/S = 1.0

The primary side powers are given as following. ∗

= PA + jQA = V A I A = 230∠0◦ × 99.59∠ − 30◦

SA

= 22905.7∠ − 30◦ = 19837.50 − j11453.16 VA

PA

= 19837.50 W, QA = −11453.160 VAr ∗

= PB + jQB = V B I B = 230∠ − 120◦ × (−99.59∠30◦ )∗

SB

= 22905.7∠30◦ = 19837.50 + j11453.16 VA PB

= 19837.50 W, QB = 11453.160 VAr

SC PC

= PC + jQC = V C IC = 230∠120◦ × 0



= 0 W, QC = 0 VAr

The total active and reactive powers on the primary side are given as follows. P

= PA + PB + PC = 39675 W

Q

= QA + QB + QC = 0 VAr   = SA + SB + SC  = P = 39675 VA       = SA  + SB  + SC  = 22906 + 22906 + 0 = 45813 VA = 3Ve Ie = 3 ∗ 230 ∗ 81.32 = 56109 VA

Svect SArith Se f f p fvect p fArith p fe

= P/Svect = 1.0

= P/SArith = 39675/45813 = 0.866 = P/Se = 39675/45813 = 0.707

Control Theories for Load Compensation

233

Example 5.3. Consider an inductive load connected as shown in Fig. 5.7. It is supplied from a symmetrical source of a sinusoidal balanced voltage with va = √ 2V sin ωt, with V = 230 Volts, 50 Hz. Express the voltage and currents for the primary and secondary side of the transformer. Express the active and reactive components of the currents, powers and discuss them.

Figure 5.8 An unbalanced reactive load supplied by three-phase delta-star connected transformer

Solution: With the above given values, the primary side phase voltages with respect to the virtual ground could be expressed as the following. √ √ vA = 2V sin ωt = 230 2 sin ωt √ √ vB = 2V sin(ωt − 120◦ ) = 230 2 sin(ωt − 120◦ ) (5.105) √ √ ◦ ◦ vC = 2V sin(ωt + 120 ) = 230 2 sin(ωt + 120 ) Therefore, the primary side line-to-line voltages are expressed as follows. √ √ 2 3V sin(ωt + 30◦ ) vAB = √ √ vBC = 2 3V sin(ωt − 90◦ ) √ √ vCA = 2 3V sin(ωt + 150◦ ) These voltages are transformed to the secondaries and are expressed below. √ √ √ 2 3V sin ωt = 398.37 2 sin(ωt + 30◦ ) va = √ √ √ vb = 2 3V sin(ωt − 120◦ ) = 398.37 2 sin(ωt − 90◦ ) √ √ √ vc = 2 3V sin(ωt + 120◦ ) = 398.37 2 sin(ωt + 150◦ ) In phasor form, the above voltages are expressed below. √ Va = 3V ∠30◦ V √ Vb = 3V ∠ − 90◦ V √ Vc = 3V ∠150◦ V

234

Power Quality in Power Distribution Systems: Concepts and Applications

Currents on the secondary side are calculated as. √ √ √ 2 3V sin(ωt − 60◦ ) = 99.59 2 sin(ωt − 60◦ ) ia = X ib = 0 ic

= 0

In phasor form, the above can be expressed as, I a = 99.59∠ − 60◦ A. The above phase-a current (ia ) is transformed to the primary of the delta-connected winding. Since the currents should have a 90◦ phase shift with respect to the voltages across the windings as given by (5.84), therefore the currents on the primary side of the transformer are given as follows. √ iA = iAB = ia = 2I sin(ωt − 60◦ ) √ iB = −iA = − 2I sin(ωt − 60◦ ) iC

= 0

In phasor form, the above can be expressed as, IA IB IC

= I∠ − 60◦ = 99.59∠ − 60◦ A

= −I∠ − 60◦ = −99.59∠ − 60◦ A = 0 A.

After knowing the voltages and currents of the primary side of the transformer, their α and β components are expressed as.       3 0 vα vA 2   √ = vβ √1 2 vB 2

Substituting vA and vB from (5.105) in the above equation, we get the following.   √   √     3 0 vα 2V sin ωt 3V sin ωt √ = =  12 √  √ (5.106) vβ − 3V cos(ωt) 2V sin(ωt − 120◦ ) √ 2 2

and,

   3 iα =  12 iβ √ 2

      √ 3 0  iA 0 2 I sin(ωt − 60◦ ) 2   √ √ √ = ◦ √1 2 iB 2 − 2 I sin(ωt − 60 ) 2 √  3 I sin(ωt − 60◦ ) = (5.107) −I sin(ωt − 60◦ )

235

Control Theories for Load Compensation

Based on the above, the active and reactive powers are computed as. p(t) = pαβ

= =

= = = q(t)

= = = = = =

vα iα + vβ iβ √ √ 3V sin(ωt) 3I sin(ωt − 60◦ ) √ +(− 3V cos ωt)(−I sin(ωt − 60◦ ))  √ √ 1 3 ◦ 2 3V I sin(ωt − 60 ) sin ωt + cos ωt 2 2 √ 3V I [2 sin(ωt − 60◦ ) cos(ωt − 60◦ )] √ 3V I sin 2(ωt − 60◦ )

(5.108)

vα iβ − vβ iα √ √ √ 3V sin ωt {−I sin(ωt − 60◦ )} − (− 3V cos ωt) 3I sin(ωt − 60◦ )   √ √ 3 1 ◦ sin ωt − cos ωt − 3V I 2 sin(ωt − 60 ) 2 2 √ − 3V I 2 sin(ωt − 60◦ )(sin(ωt − 60◦ )) √ − 3V I 2 sin2 (ωt − 60◦ ) √ − 3V I {1 − cos 2(ωt − 60◦ )} (5.109)

Based on above values of p and q powers, the α and β components of active and reactive components are given below.     iα p + iαq iα = iβ iβ p + iβ q where, iα p

=

vα p v2α + v2β



3V sin ωt √ √ p 2 ( 3V sin ωt) + (− 3V cos ωt)2 √ 1 = √ sin ωt 3V I {sin 2(ωt − 60◦ )} 3V = I sin ωt sin 2(ωt − 60◦ )

=

(5.110)

236

Power Quality in Power Distribution Systems: Concepts and Applications

Similarly, iβ p

=

v2α

vβ p + v2β

√ − 3V cos ωt √ √ = p ( 3V I sin ωt)2 + (− 3V cos ωt)2 √ − 3V cos ωt √ = 3V I sin 2(ωt − 60◦ ) 3V 2 = −I cos ωt sin 2(ωt − 60◦ ) iαq

−vβ q + v2β √  −(− 3V cos ωt)  √ = − 3V I [1 − cos 2(ωt − 60◦ )] 2 3V = −I cos ωt {1 − cos 2(ωt − 60◦ )} =

iβ q

(5.111)

v2α

vα q v2α + v2β √  3V sin ωt  √ ◦ = − 3V I [1 − cos 2(ωt − 60 )] 3V 2 = −I sin ωt {1 − cos 2(ωt − 60◦ )}

(5.112)

=

(5.113)

Thus, knowing iα p , iαq , iβ p and iβ q , we can determine active and reactive components of currents on the source side as given below.      −1 iα p iAp = C iβ p iBp

where,

[C]

−1



=

3 2

√1 2

−1  2 0  √ = −13 √ 2 6

0 √1 2



(5.114)

Using the above equation, we can find out the active and reactive components of the current, as given below.      2 0 I sin ωt sin 2(ωt − 60◦ ) iAp 3 = −1 iBp −I cos ωt sin 2(ωt − 60◦ ) √ √1 6

2

Control Theories for Load Compensation

237

From the above,  2 iAp = I sin ωt sin 2(ωt − 60◦ ) 3  2 I {2 sin ωt sin 2(ωt − 60◦ )} = 32 I = √ {cos(ωt − 120◦ ) − cos(3ωt − 120◦ )} 6 1 1 iBp = − √ iα p + √ iβ p 6 2 1 1 = − √ I sin ωt {sin 2(ωt − 60◦ )} + √ (−I cos ωt) {sin 2(ωt − 60◦ )} 6 2   √ I 3 1 = − √ {2 sin 2(ωt − 60◦ )} sin ωt + cos ωt 2 2 6 I = √ {cos ωt + cos(3ωt − 60◦ )} 6  2 iαq iAq = 3  2 I {1 − cos 2(ωt − 60◦ )} cos ωt = − 3 I = − √ {2 cos ωt − 2 cos ωt cos 2(ωt − 60◦ )} 6 I = − √ {2 cos ωt − cos(3ωt − 120◦ ) − cos(ωt − 120◦ )} 6 I = √ {−2 cos ωt + cos(ωt − 120◦ ) + cos(3ωt − 120◦ )} 6 1 1 iBq = − √ iαq + √ iβ q 6 2 1 = − √ {−I cos ωt [1 − cos 2(ωt − 60◦ )]} 6 1 + √ {−I sin ωt [1 − cos 2(ωt − 60◦ )]} 2   √ 2I 3 1 = √ {1 − cos 2(ωt − 60◦ )} cos ωt − sin ωt 2 2 6 = = =

2I √ {1 − cos 2(ωt − 60◦ )} cos(ωt + 60◦ ) 6 I √ {2 cos(ωt + 60◦ ) − 2 cos(ωt + 60◦ ) cos 2(ωt − 60◦ )} 6 I √ {2 cos(ωt + 60◦ ) + cos ωt − cos(3ωt − 60◦ )} 6

238

Power Quality in Power Distribution Systems: Concepts and Applications

Thus, we have, iAp

=

iBp

=

iAq

=

iBq

=

I √ {cos(ωt − 120◦ ) − cos(3ωt − 120◦ )} 6 I √ {cos ωt + cos(3ωt − 60◦ )} 6 I √ {−2 cos ωt + cos(ωt − 120◦ ) + cos(3ωt − 120◦ )} 6 I √ {cos ωt + 2 cos(ωt + 60◦ ) − cos(3ωt − 60◦ )} 6

(5.115) (5.116) (5.117) (5.118)

From the above equations, it is clear that there exist active components of currents (iAp , iBp ), even though there is a purely reactive load. Additionally, both active and reactive components of currents (iAp , iAq , iBp , iBq ), have third harmonics, although there is purely linear inductive load. Thus, this does not go in line with the definitions of active and reactive components of currents proposed by Fryze given in [26]. Powers computation The secondary side powers are given as follows. ∗

Sa Pa

= Pa + jQa = V a I a = 398.37∠30◦ 99.59∠60◦ = 39673.8361∠90◦ VA

Sb Pb Sc Pc

= Pb + jQb = V b I b = 398.37∠ − 90◦ × 0 = 0 VA

= 0 W, Qa = 39673.8361 VAr ∗

= 0 W, Qb = 0 VAr ∗ = Pc + jQc = V c I c = 398.37∠150◦ × 0 = 0 VA = 0 W, Qc = 0 VAr

The total active and reactive powers on the secondary side are given as follows. P

= Pa + Pb + Pc = 0 W

Q

= Qa + Qb + Qc = 39673.83 VAr

Svect p fvect

= Sarith = Q = 39673.83 VA =

p farith = P/S = 0

The primary side powers are given as follows. ∗

SA PA SB

= PA + jQA = V a I a = 230∠0◦ 99.59∠60◦ = 11452.85 + j19836.91 VA = 11452.85 W, QA = 19836.91 VAr

PB

= −11452.85 W, QB = 19836.91 VAr

SC PC



= PB + jQB = V b I b = 230∠ − 120◦ (−99.59∠ − 120◦ )∗ = ∗

= PC + jQC = V C IC = 230∠120◦ × 0 = 0 W, QC = 0 VAr

Control Theories for Load Compensation

239

The total active and reactive powers on the primary side are given as follows. P = PA + PB + PC = 0 W Q = QA + QB + QC = 39673.82 VAr   Svect = SA + SB + SC  = 39673.82 VA       SArith = SA  + SB  + SC  = 22906 + 22906 + 0 = 45811.4 VA Se = 3Ve Ie = 3 ∗ 230 ∗ 81.32 = 56109 VA p fvect = P/Svect = 0 p fArith

=

P/SArith = 0

p fe

=

P/Se = 39675/45813 = 0

Example 5.4. Consider the star-delta connected ideal transformer with 1:1 turns ratio, as shown in Fig. 5.9. The secondary side of transformer, a load of 3 ohms is connected between the phase-a and b. The primary is star connected and neutral is grounded. Compute the following. (a) Draw the circuit diagram and compute time domain expressions of currents in each phase on both primary and secondary side. (b) Does the load require reactive power from the source? If any, find its value. Also, compute the reactive power in each phase on either sides of the transformer. (c) Also, determine active powers in each phase and the overall active power on either sides of the transformer. (d) If you have a similar arrangement with balanced load with same output power, comment upon the rating of the line conductors and the transformer.

Figure 5.9 An unbalanced load supplied by three-phase star-delta connected transformer

Solution: In this example, we have a three-phase star-delta connected transformer of turns ratio 1:1 with the star side connected to a three-phase balanced voltage source and the neutral connected to the ground. Thus, in this three-phase star-delta connected transformer, the delta side phase voltages equal the star side line voltages.

240

Power Quality in Power Distribution Systems: Concepts and Applications

The primary side instantaneous phase voltages are given by, √ vA = 230 2 sin ωt = 325.27 sin ωt √ vB = 230 2 sin(ωt − 120◦ ) = 325.27 sin(ωt − 120◦ ) √ ◦ vC = 230 2 sin(ωt + 120 ) = 325.27 sin(ωt + 120◦ ) Therefore, the instantaneous line to line voltages on the delta side (secondary) are given by, √ vab = 230 2 sin ωt = 325.27 sin ωt √ vbc = 230 2 sin(ωt − 120◦ ) = 325.27 sin(ωt − 120◦ ) √ vca = 230 2 sin(ωt + 120◦ ) = 325.27 sin(ωt + 120◦ ) Therefore, the instantaneous phase voltages with respect to the ground are given as follows. √ √ 230 2 √ va = sin(ωt − 30◦ ) = 132.79 2 sin(ωt − 30◦ ) 3 √ √ 230 2 √ vb = sin(ωt − 150◦ ) = 132.79 2 sin(ωt − 150◦ ) 3 √ √ 230 2 √ vc = sin(ωt + 90◦ ) = 132.79 2 sin(ωt + 90◦ ) 3 (a) On the delta side, we have a resistive load of R = 3 Ω connected between the terminals a and b. Thus, the expression for instantaneous currents flowing out of terminals a, b, and c of the transformer are given by, √ √ 230 2 vab = sin ωt = 76.67 2 sin ωt ia = R 3 √ √ vab 230 2 =− sin ωt = −76.67 2 sin ωt ib = −ia = − R 3 ic = 0 Therefore, the winding currents on the secondary side are given as. √ iab = 76.67 2 sin ωt ibc = 0 ica

= 0

These currents are transformed to the primary windings. Thus, the time domain expressions of these currents are given as, √ iA = 76.67 2 sin ωt iB = 0 iC

= 0

241

Control Theories for Load Compensation

(b) Load does not require any reactive power from the source because it is a purely resistive load. This fact can also be verified by looking at the expressions for instantaneous phase voltages and currents on the star side. (c) Similarly, no current (and hence power) is drawn by the load from phases-B and C. Thus, various powers on the primary side are as follows. For phase-A, SA PA



= PA + jQA = V A I A = 230∠0◦ × 76.67∠0◦ = 17634.1 VA = 17634.1 W, QA = 0 VAr

For phase-B, SB PB



= PB + jQB = V B I B = 230∠ − 120◦ × 0 = 0 VA = 0 W, QB = 0 VAr

For phase-C, SC PC



= PC + jQC = V C IC = 230∠120◦ × 0 = 0 VA = 0 W, QC = 0 VAr

Thus, various powers on the secondary side are as follows. For phase-A, Sa

= =

Pa

=

For phase-B,



Pa + jQa = V a I a = 132.79∠ − 30◦ × 76.67∠0◦ 8817.05 − j5090.53 VA

8817.05 W, Qa = −5090.53 VAr ∗

Sb

= Pb + jQb = V b I b = 132.79∠ − 150◦ × (−76.67∠0◦ ) = 8817.05 + j5090.53 VA

Pb

=

8817.05 W, Qb = 5090.53 VAr

For phase-C, Sc

=

Pc

=

Pc + jQc = V c I c = 132.79∠90◦ × 0 = 0 VA



0 W, Qc = 0 VAr

For the above analysis, it is observed that the total active power, P = PA + PB + PC = Pa + Pb + Pc = 17634.1 W and the total reactive power Q = QA + QB + QC = Qa +Qb +Qc = 0 VAr. However, due to the unbalanced load, on the delta side of the transformer phase-a and phase-b experience reactive powers as calculated above. This causes a non-unity power factor in phase-a and b. The vector and arithmetic apparent powers are: Svect = |Sa +Sb +Sc | = 17634.1 VA and SArith = |Sa |+|Sb |+|Sc | = 17634.1 VA, and Se = 3Ve Ie = 3∗230∗62.59 = 41393 VA. Accordingly the vector, arithmetic, and effective power factors are 1.0, 1.0, and 0.4082, respectively. (d) Earlier, the load which was getting power from one phase on the delta side, will now be shared equally among the three phases. Thus, the current rating of the line conductors and transformer will reduce.

242

5.4

Power Quality in Power Distribution Systems: Concepts and Applications

THEORY OF INSTANTANEOUS SYMMETRICAL COMPONENTS

The theory of instantaneous symmetrical components can be used for the purpose of load balancing, harmonic suppression, and power factor correction [7]. The control algorithms based on instantaneous symmetrical component theory can practically compensate any kind of unbalance and harmonics in the load, provided we have a high bandwidth power converter to track the filter reference currents. These algorithms are discussed in this section. For any set of three-phase instantaneous voltages, the instantaneous symmetrical components are given as.      1 1 1 va0 va 1 va+  = 1 a a2  vb  (5.119) 3 va− vc 1 a2 a Similarly, three-phase instantaneous currents can be expressed in their instantaneous symmetrical components, as given in the following.      1 1 1 ia0 ia 1 ia+  = 1 a a2  ib  (5.120) 3 ic ia− 1 a2 a

In the above equations, a is a complex operator and it is given by a = e j 2π/3 and a2 = e j 4π/3 . It is to be noted that the instantaneous components of currents, ia+ and ia− are complex time varying quantities and also they are complex conjugates of each other. The same is true for va+ and va− quantities. The terms ia0 and va0 are real quantities, however (-) has been used as upper script for the sake of uniformity of notation. These instantaneous symmetrical components are used to formulate equations for load compensation. First, a three-phase, four-wire system supplying star-connected load is considered. 5.4.1

COMPENSATING STAR CONNECTED LOAD

A three-phase four-wire compensated system is shown in Fig. 5.10. In the figure, three-phase load currents (ila , ilb and ilc ), can be unbalanced and nonlinear load. The objective in either three or four-wire compensation system is to provide a balanced supply current such that its zero sequence component is zero. We therefore have, isa + isb + isc = 0

(5.121)

Using equations (5.119) and (5.120), instantaneous positive sequence voltage (va+ ) and current (ia+ ) are computed from instantaneous values of vsa , vsb , vsc and isa , isb , isc , respectively. To have a predefined power factor from the source, the relationship between the angle of va+ and ia+ is given as follows. ∠vsa+

= ∠isa+ + φ+

where φ+ is desired phase angle between va+ and ia+ . The above equation is rewritten as follows.     1 1 ∠ (vsa + a vsb + a2 vsc ) = ∠ (isa + a isb + a2 isc ) + φ+ 3 3

(5.122)

Control Theories for Load Compensation

243

Figure 5.10 A three-phase four-wire compensated system

L.H.S. of the above equation is expressed as below.     √  √   1 1 3 3 1 vsa + − + j vsb + − − j vsc L.H.S. = ∠ 3 2 2 2 2    √ vsb vsc  3 1  vsa − +j − (vsb − vsc ) = ∠ 3 2 2 2 √ ( 3/2) (vsb − vsc ) = tan−1 (vsa − vsb /2 − vsc /2) K1 (5.123) = tan−1 K2 √ where, K1 = ( 3/2) (vsb − vsc ) and K2 = (vsa − vsb /2 − vsc /2). Similarly, R.H.S. of the equation is expanded as below.     √  √   3 3 1 1 1 + φ+ isa + − + j isb + − − j isc R.H.S. = ∠ 3 2 2 2 2    √  3 1 isb isc isa − − +j (isb − isc ) = ∠ + φ+ 3 2 2 2 √ ( 3/2) (isb − isc ) + φ+ = tan−1 (isa − isb /2 − isc /2) K3 = tan−1 + φ+ (5.124) K4 √ where, K3 = ( 3/2) (isb − isc ) and K4 = (isa − isb /2 − isc /2). Equating (5.123) and (5.124), we get the following. tan−1

K1 K3 = tan−1 + φ+ K2 K4

244

Power Quality in Power Distribution Systems: Concepts and Applications

Taking tangent on both sides, the following is obtained.     K1 K3 = tan tan−1 tan tan−1 + φ+ K2 K4 (K3 /K4 ) + tan φ+ K1 = Therefore, K2 1 − (K3 /K4 ) × tan φ+ K1 K3 + K4 tan φ+ = K2 K4 − K3 × tan φ+ The above equation implies that, K1 K4 − K1 K3 tan φ+ − K2 K3 − K2 K4 tan φ+ = 0 Substituting the values of K1 , K2 , K3 , K4 in the above equation, the following expression is obtained. √   3 3 isb isc (vsb − vsc ) isa − − − (vsb − vsc ) (isb − isc ) tan φ+ 2 2 2 4 √  3 vsb vsc  vsa − − (isb − isc ) − 2 2 2    vsb vsc isb isc isa − tan φ+ = 0 − vsa − − − 2 2 2 2 The above equation can be arranged according to the terms associated with isa , isb , and isc as given below. √  3 tan φ+ (vsb − vsc ) + (vsb + vsc − 2 vsa ) isa 2 2 √  tan φ+ 3 + (vsc − vsa ) + (vsc + vsa − 2 vsb ) isb 2 2 √  tan φ+ 3 + (vsa − vsb ) + (vsa + vsb − 2 vsc ) isc = 0 2 2 √

Dividing above equation by 23 , it can be written as follows.   tan φ+ (vsb − vsc ) + √ (vsb + vsc − 2 vsa ) isa 3   tan φ+ + (vsc − vsa ) + √ (vsc + vsa − 2 vsb ) isb 3   tan φ+ + (vsa − vsb ) + √ (vsa + vsb − 2 vsc ) isc = 0 3

Control Theories for Load Compensation

√ Let tan φ+ / 3 = β , the above equation is further simplified to, {(vsb − vsc ) + β (vsb + vsc − 2 vsa )} isa + {(vsc − vsa ) + β (vsc + vsa − 2 vsb )} isb

245

+ {(vsa − vsb ) + β (vsa + vsb − 2 vsc )} isc = 0.

(5.125)

ps = vsa isa + vsb isb + vsc isc = Plavg

(5.126)

The third objective of compensation is that the power supplied from the source (ps ) must be equal to the average load power (Plavg ). Therefore the following holds true. The above equation has important implications. For example, when supply voltages are balanced, the equation is satisfied for balanced source currents. However, if supply voltages are unbalanced and distorted, the equation gives a set of currents that are also not balanced and sinusoidal in order to supply constant power. Equations (5.121), (5.125), and (5.126), can be written in matrix form as given below.   1 1 1   isa  (vsb − vsc )  (v − v ) (v − v ) sc sa sa sb      isb  +β (vsb + vsc − 2vsa ) +β (vsc + vsa − 2vsb ) +β (vsa + vsb − 2vsc )  isc vsb vsc vsa   0 = 0  Plavg

which can be further written as,      A isabc = Plavg

(5.127)

Therefore,

     isabc = A−1 Plavg

=

=



T   ac11 ac12 ac13 0 1  ac21 ac22 ac23   0  |A| ac31 ac32 ac33 Plavg    ac11 ac21 ac31 0 1  ac12 ac22 ac32   0  |A| ac13 ac23 ac33 Plavg

where aci j is the cofactor of ith row and jth column element of matrix A in (5.127) and |A| is the determinant of matrix A. Due to the presence of zero elements in the first two rows of the column matrix with power elements, the cofactors in the first two columns need not be computed. These are indicated by dots in the following matrix.        · · ac31 ac31 0 isa   1 1 · · ac32   0  = ac32  Plavg isabc = isb  = |A| |A| isc · · ac33 Plavg ac33

246

Power Quality in Power Distribution Systems: Concepts and Applications

The determinant of matrix A is computed as below. |A| = [(vsc − vsa ) + β (vsc + vsa − 2vsb )]vsc

−[(vsa − vsb ) + β (vsa + vsb − 2vsc )]vsb −[(vsb − vsc ) + β (vsb + vsc − 2vsa )]vsc

+[(vsa − vsb ) + β (vsa + vsb − 2vsc )]vsa +[(vsb − vsc ) + β (vsb + vsc − 2vsa )]vsb −[(vsc − vsa ) + β (vsc + vsa − 2vsb )]vsa

= β [v2sc + vsa vsc − 2vsb vsc − vsa vsb − v2sb + 2vsb vsc − vsb vsc − v2sc + 2vsa vsc

+v2sa + vsb vsa − 2vsc vsa + v2sb + vsc vsb − 2vsa vsb − vsc vsa − v2sa + 2vsb vsa ]

+(vsc − vsa )vsc − (vsa − vsb )vsb − (vsb − vsc )vsc

+(vsa − vsb )vsa + (vsb − vsc )vsb − (vsc − vsa )vsa

The above equation can be further simplified to, |A| = β · 0 + v2sc − vsa vsc − vsb vsa + v2sb − vsb vsc + v2sc + v2sa −vsa vsb + v2sb − vsc vsb − vsa vsc + v2sa

= 2v2sa + 2v2sb + 2v2sc − 2vsa vsb − 2vsb vsc − 2vsc vsa

= v2sa + v2sb − 2vsa vsb + v2sb + v2sc − 2vsb vsc + v2sc + v2sa − 2vsc vsa = =

(vsa − vsb )2 + (vsb − vsc )2 + (vsc − vsa )2 (v2sab + v2sbc + v2sca )

Further adding and subtracting, v2sa + v2sb + v2sc in the above equation, we get the following. |A| = 3(v2sa + v2sb + v2sc ) − (v2sa + v2sb + v2sc + 2vsa vsb + 2vsb vsc + 2vsc vsa ) (5.128) Further, (vsa + vsb + vsc )2

= v2sa + v2sb + v2sc + 2vsa vsb + 2vsb vsc + 2vsc vsa (5.129)

Using equations (5.128) and (5.129), we obtain the following. |A| = 3(v2sa + v2sb + v2sc ) − (vsa + vsb + vsc )2 = 3



j=a,b,c

= 3





v2s j − 9v2so

j=a,b,c

v2s j − 3v2so



(5.130)

In above equation, the term vso is the instantaneous zero sequence component of the source voltage and it is given as following. vso =

(vsa + vsb + vsc ) 3

(5.131)

Control Theories for Load Compensation

247

The determinant of matrix A, using (5.130), can also be expressed as, |A| = =

= = =

  3 v2sa + v2sb + v2sc − 3 v2so   3 v2sa + v2sb + v2sc + 3v2so − 2 × 3v2so

3[v2sa + v2sb + v2sc + v2so + v2so + v2so − 2vso (vsa + vsb + vsc )]

3[(v2sa + v2so − 2vsa vso ) + (v2sb + v2so − 2vsb vso ) + (v2sc + v2so − 2vsc vso )] 3[(v2sa − v2so ) + (v2sb − v2so ) + (v2sc − v2so )]

The cofactors ac31 , ac32 , and ac33 are computed as below. ac31

= [−(vsc − vsa ) − β (vsc + vsa − 2vsb ) + (vsa − vsb ) + β (vsa + vsb − 2vsc )] = vsa − vsb − vsc + vsa + β (−vsc − vsa + 2vsb + vsa + vsb − 2vsc ) = (2vsa − vsb − vsc ) + 3β (vsb − vsc ) = (2vsa + vsa − vsa − vsb − vsc ) + 3β (vsb − vsc )

=

3 (vsa − vs0 ) + 3 β (vsb − vsc )

Similarly, ac32

= [−(vsa − vsb ) − β (vsa + vsb − 2vsc ) + (vsb − vsc ) + β (vsb + vsc − 2vsa )] = (−vsa + vsb + vsb − vsc ) + β (−vsa − vsb + 2vsc + vsb + vsc − 2vsa ) = (2vsb − vsc − vsa ) + 3β (vsc − vsa ) =

3(vsb − vso ) + 3β (vsc − vsa )

and, ac33

= [(vsc − vsa ) + β (vsc + vsa − 2vsb ) − (vsb − vsc ) − β (vsb + vsc − 2vsa )] = (2vsc − vsa − vsb ) + 3β (vsa − vsb )

=

3(vsc − vso ) + 3β (vsa − vsb )

Knowing the value of cofactors, we now have,     isa 3(vsa − vs0 ) + 3β (vsb − vsc )   1 isb  =  3(vsb − vs0 ) + 3β (vsc − vsa ) Plavg  3 ∑ j=a,b,c v2s j − 3v2so 3(vsc − vs0 ) + 3β (vsa − vsb ) isc   (vsa − vso ) + β (vsb − vsc )   1  (vsb − vso ) + β (vsc − vsa ) Plavg (5.132) =  ∑ j=a,b,c v2s j − 3v2so (vsc − vso ) + β (vsa − vsb )

248

Power Quality in Power Distribution Systems: Concepts and Applications

From the above equation, the desired source currents can be written as following. isa

=

isb

=

isc

=

(vsa − vso ) + β (vsb − vsc ) Plavg ∑ j=a,b,c v2s j − 3 v2so (vsb − vso ) + β (vsc − vsa ) Plavg ∑ j=a,b,c v2s j − 3 v2so

(5.133)

(vsc − vso ) + β (vsa − vsb ) Plavg ∑ j=a,b,c v2s j − 3 v2so

Applying Kirchoff’s current law at the point of common coupling (PCC), we have, i∗f a i∗f b i∗f c

= ila − isa = ilb − isb

(5.134)

= ilc − isc

Replacing isa , isb , and isc from equations (5.133)–(5.134), we obtain the reference filter currents as given in the following. i∗f a

= ila − isa = ila −

i∗f b

= ilb − isb = ilb −

i∗f c

= ilc − isc = ilc −

(vsa − vso ) + β (vsb − vsc ) Plavg ∑ j=a,b,c v2s j − 3 v2so (vsb − vso ) + β (vsc − vsa ) Plavg ∑ j=a,b,c v2s j − 3 v2so

(5.135)

(vsc − vso ) + β (vsa − vsb ) Plavg ∑ j=a,b,c v2s j − 3 v2so

For balanced supply voltages, vs0 = 0. For the unity power factor, β = 0. Therefore, expressions for the reference filter currents are simplified to the following. i∗f a i∗f b i∗f c

vsa Plavg ∑ j=a,b,c v2s j vsb = ilb − isb = ilb − Plavg ∑ j=a,b,c v2s j vsc = ilc − isc = ilc − Plavg ∑ j=a,b,c v2s j = ila − isa = ila −

(5.136)

Example 5.5. Consider a three-phase four-wire compensated system shown in Fig. 5.10. The three-phase voltages and load impedances as follows. √ vsa (t) = 230 2 sin ω t √ vsb (t) = 230 2 sin(ω t − 120◦ ) √ vsc (t) = 230 2 sin(ω t + 120◦ ) Load impedances: Zla = 5 + j 12 Ω, Zlb = 3 + j 4 Ω, Zlc = 5 − j 12 Ω. Using instantaneous symmetrical component theory,

249

Control Theories for Load Compensation

(a) Find reference filter currents (i f a , i f b , i f c ) of an ideal compensator such that source currents are balanced sinusoids and in phase with their respective source voltages. Also, find the active and reactive powers of the compensator. (b) Based on these reference filter currents, identify the nature and values of Z f a , Z f b , Z f c . Are these practically realizable using passive impedances? (c) Find out the time expressions for the source currents based on the above filter currents. Check whether these are balanced and in phase with their respective voltages. (d) Compute the source powers and check if Ps = Pl , Qs = 0, Pf = 0, and Q f = Ql . Solution: (a) Given the source voltage and load impedances, the load currents are computed as following. I la

=

I lb

=

I lc

=

V sa 230∠0◦ = 17.69∠ − 67.38◦ A = Zla 5 + j 12 V sb 230∠ − 120◦ = 46.0∠ − 173.13◦ A = Zlb 3+ j4 V sc 230∠120◦ = 17.69∠187.38◦ A = Zlc 5 − j 12

For the given source voltages, the value of v2sa + v2sb + v2sc is given below. 2 = 3 × 2302 = 158700 Volt2 v2sa + v2sb + v2sc = 3Vrms

The load real and reactive powers are computed as following. ∗





Sl = V sa I la +V sb I lb +V sc I lc = 9478.2 + j 8464 VA This gives, Pl = Plavg = 9478.2 W, and Ql = 8464 VAr. The reference filter currents are computed using (5.136). Since the voltage and current quantities involved here have only fundamental components, the reference filter currents can be computed using phasor equations equivalent to (5.136). These are expressed below. I fa I fb I fc

Plavg v2sa + v2sb + v2sc Plavg = I lb −V sb 2 vsa + v2sb + v2sc Plavg = I lc −V sc 2 vsa + v2sb + v2sc = I la −V sa

250

Power Quality in Power Distribution Systems: Concepts and Applications

Substituting the values of source voltages, load currents, Plavg , and v2sa + v2sb + v2sc as computed above in (5.137), we get the following. I fa I fb I fc

9478.2 = 17.74∠ − 113◦ A 158700 9478.2 = 39.32∠170.64◦ A = 46.0∠ − 173.13◦ − 230∠ − 120◦ 158700 9478.2 = 17.69∠187.38◦ − 230∠120◦ = 17.74∠ − 127◦ A 158700 = 17.69∠ − 67.38◦ − 230∠0

Therefore, the time expressions for filter currents are given as following. √ i f a (t) = 17.74 2 sin(ω t − 113◦ ) A √ i f b (t) = 39.32 2 sin(ω t + 170.64◦ ) A √ i f c (t) = 17.74 2 sin(ω t − 127◦ ) A Based on the above reference filter currents, the compensator powers are computed as below.  ∗ ∗ ∗  Pf = Real V sa I f a +V sb I f b +V sc I f c = 0 W  ∗ ∗ ∗  Q f = Imag V sa I f a +V sb I f b +V sc I f c = 8464 VAr = Ql

From the values of compensator real and reactive powers, it can be observed that the compensator does not exchange any real power with the load or source, and it supplies the required reactive power to the load. Thus, the source is relieved of supplying reactive power of the load and maintains a unity power factor.

(b) Based on the voltages at the load point and filter currents, the compensator impedances Z f a , Z f b , Z f c are computed as following. Z fa

=

Z fb

=

Z fc

=

V sa 230∠0◦ = = 5.06 − j 11.93 Ω −17.74∠ − 113◦ −I f a

230∠ − 120◦ V sb = = −2.06 − j 5.47 Ω −39.32∠170.64◦ −I f b V sc 230∠120◦ = = 5.06 + j 11.93 Ω −17.74∠ − 127◦ −I f c

It can be seen that the compensator impedances in leg-a and c can be realized as capacitive and inductive loads, respectively. However, leg-b has a negative real part in the impedance and hence can not be realized using a passive component. Thus, the scheme can not be implemented using a passive filter. That is why the idea of active filtering is employed to realize such compensation using a voltage source inverter.

Control Theories for Load Compensation

251

(c) The compensated source currents are calculated below. I sa I sb I sc

= I la − I f a = 13.74 ∠0 A

= I lb − I f b = 13.74 ∠ − 120◦ A = I lc − I f c = 13.74 ∠120◦ A

Therefore, the time expressions of the source currents are given as following. √ isa (t) = 13.74 2 sin ω t A √ isb (t) = 13.74 2 sin(ω t − 120◦ ) A √ isc (t) = 13.74 2 sin(ω t + 120◦ ) A From the above, it can be observed that the compensated source currents are balanced and in phase with their respective source voltages. (d) The source and compensator powers are given as below.  ∗ ∗ ∗  Ps = Pl = Real V sa I sa +V sb I sb +V sc I sc = 9478.2 W  ∗ ∗ ∗  Qs = Imag V sa I sa +V sb I sb +V sc I sc = 0 VAr Pf = 0 W Q f = Ql = 8464 VAr It can be observed that the source only supplies the real power of the load, and the compensator supplies the total reactive power of the load. 5.4.2

COMPENSATING DELTA CONNECTED LOAD

The balancing of an unbalanced ∆-connected load is a generic problem and the theory of instantaneous symmetrical components can be used to balance the load. The schematic diagram of this compensated scheme is shown in Fig. 5.11.

Figure 5.11 A compensation for delta connected load

252

Power Quality in Power Distribution Systems: Concepts and Applications

This compensator is connected between the phases of the load. The aim is to generate the three reference current waveforms denoted by i∗f ab , i∗f bc , and i∗f ca , respectively based on the measurement of system voltages and load currents such that the supply sees a balanced load. As there is no fourth or neutral wire, the following equation is valid. isa + isb + isc = 0

(5.137)

Applying Kirchhoff’s current law at nodes, we can express isa , isb , and isc , respectively as follows. isa

=

isb

=

isc

=

(ilab − i∗f ab ) − (ilca − i∗f ca ) (ilbc − i∗f bc ) − (ilab − i∗f ab )

(5.138)

(ilca − i∗f ca ) − (ilbc − i∗f bc )

As can be seen from the above equations, (5.137) is satisfied. Since, in a balanced ∆connected load after compensation, the zero sequence current within the delta cannot flow, therefore, (ilab − i∗f ab ) + (ilbc − i∗f bc ) + (ilca − i∗f ca ) = 0

(5.139)

The source supplies the average load power, Plavg and the following equation is satisfied. vsa isa + vsb isb + vsc isc = Plavg

(5.140)

From the power factor condition between the source voltages and currents, as stated in (5.122), the following equation was derived. {(vsb − vsc ) + β (vsb + vsc − 2 vsa )} isa

+ {(vsc − vsa ) + β (vsc + vsa − 2 vsb )} isb + {(vsa − vsb ) + β (vsa + vsb − 2 vsc )} isc = 0.

(5.141)

Replacing isa , isb and isc from (5.138) in the above equation, the following equation is obtained. {[(vsb − vsc ) + β (vsb + vsc − 2vsa )] − [(vsc − vsa ) + β (vsc + vsa − 2vsb )]} (ilab − i∗f ab )    I

+ {[(vsc − vsa ) + β (vsc + vsa − 2vsb )] − [(vsa − vsb ) + β (vsa + vsb − 2vsc )]} (ilbc − i∗f bc )    II

+ {[(vsa − vsb ) + β (vsa + vsb − 2vsc )] − [(vsb − vsc ) + β (vsb + vsc − 2vsa )]} (ilca − i∗f ca ) = 0    III

253

Control Theories for Load Compensation

The first term, I is simplified, as given below. I

= = =

{(vsb − vsc − vsc + vsa ) + β (vsb + vsc − 2vsa − vsc − vsa + 2vsb )} (ilab − i∗f ab )

{(vsa + vsb − 2vsc ) − 3β (vsa − vsb )} (ilab − i∗f ab ) −3 {(vsc − vs0 ) + β (vsa − vsb )} (ilab − i∗f ab )

(5.142)

Similarly, the second and third terms are given as below. II

= {(vsc − vsa − vsa + vsb ) + β (vsc + vsa − 2vsb − vsa − vsb + 2vsc )} (ilbc − i∗f bc ) = {(vsb + vsc − 2vsa ) − 3β (vsb − vsc )} (ilbc − i∗f bc ) = −3 {(vsa − vs0 ) + β (vsb − vsc )} (ilbc − i∗f bc ) (5.143)

III

= {(vsa − vsb − vsb + vsc ) + β (vsa + vsb − 2vsc − vsb − vsc + 2vsa )} (ilca − i∗f ca ) = {(vsc + vsa − 2vsb ) − 3β (vsc − vsa )} (ilca − i∗f ca ) = −3 {(vsb − vs0 ) + β (vsc − vsa )} (ilca − i∗f ca ) (5.144)

Summing the above three terms and simplifying we get, {(vsc − vs0 ) + β (vsa − vsb )} (ilab − i∗f ab ) + {(vsa − vs0 ) + β (vsb − vsc )} (ilbc − i∗f bc )

+ {(vsb − vs0 ) + β (vsc − vsa )} (ilca − i∗f ca ) = 0

(5.145)

The third condition for load compensation ensures that the average load power should be supplied from the sources. Therefore, vsa isa + vsb isb + vsc isc = Plavg

(5.146)

The terms isa , isb , and isc are substituted from (5.138) in the above equation, and the modified equation is given below.     vsa (ilab − i∗f ab ) − (ilca − i∗f ca ) + vsb (ilbc − i∗f bc ) − (ilab − i∗f ab )   (5.147) +vsc (ilca − i∗f ca ) − (ilbc − i∗f bc ) = Plavg The above is simplified to,

(vsa − vsb )(ilab − i∗f ab ) + (vsb − vsc )(ilbc − i∗f bc ) + (vsc − vsa )(ilca − i∗f ca ) = Plavg

(5.148)

Equations (5.139), (5.145), (5.148) can be written in the matrix form as given below.   1 1 1    ∗  0   ilab − i f ab (vsa − vs0 ) (vsb − vs0 )    (vsc − vs0 )   ilbc − i∗f bc  =  0  (5.149) +β (vsa − vsb ) +β (vsb − vsc ) +β (vsc − vsa ) Plavg ilca − i∗f ca vsb − vsc vsc − vsa vsa − vsb

254

Power Quality in Power Distribution Systems: Concepts and Applications

The above equation can be written in the following form.

Therefore,

    ilab − i∗f ab 0 [A∆ ]  ilbc − i∗f bc  =  0  Plavg ilca − i∗f ca

(5.150)

    ilab − i∗f ab 0  ilbc − i∗f bc  = [A∆ ]−1  0  Plavg ilca − i∗f ca

(5.151)

The above equation is solved by finding the determinate of A∆ and the cofactors transpose as given below.   ilab − i∗f ab  ilbc − i∗f bc  = ilca − i∗f ca =

 a 1  c11 ac21 |A∆ | ac31  a 1  c11 ac12 |A∆ | ac13

ac12 ac22 ac32 ac21 ac22 ac23

T   ac13 0 ac23   0  Plavg ac33   ac31 0 ac32   0  ac33 Plavg

(5.152)

The determinant |A∆ | and the cofactors in the above equation are calculated below. |A∆ | =

[(vsa − vso ) + β (vsb − vsc )](vsc − vsa ) − [(vsb − vso ) + β (vsc − vsa )](vsb − vsc )

− [(vsc − vso ) + β (vsa − vsb )](vsc − vsa ) + [(vsb − vso ) + β (vsc − vsa )](vsa − vsb )

+ [(vsc − vso ) + β (vsa − vsb )](vsb − vsc ) − [(vsa − vso ) + β (vsb − vsc )](vsa − vsb )

Separating all the terms containing β and rearranging the above equation, we get, |A∆ | =

(vsa − vso )(vsc − vsa ) − (vsb − vso )(vsb − vsc ) − (vsc − vso )(vsc − vsa )

+(vsb − vso )(vsa − vsb ) + (vsc − vso )(vsb − vsc ) − (vsa − vso )(vsa − vsb )

+β [(vsb − vsc )(vsc − vsa ) − (vsc − vsa )(vsb − vsc ) − (vsa − vsb )(vsc − vsa ) +(vsc − vsa )(vsa − vsb ) + (vsa − vsb )(vsb − vsc ) − (vsb − vsc )(vsa − vsb )] It is seen that the terms containing β cancel out each other and give zero. Thus, the above equation becomes, |A∆ | = (vsa − vso )(vsc − vsa − vsa + vsb )

+(vsb − vso )(−vsb + vsc + vsa − vsb )

+(vsc − vso )(−vsc + vsa + vsb − vsc ) + β × 0

(5.153)

Control Theories for Load Compensation

255

In the above, using (vsa + vsb + vsc ) = 3vs0 , vsc − vsa − vsa + vsb

=

−vsc + vsa + vsb − vsc

=

−vsb + vsc + vsa − vsb

=

vsa + vsb + vsc − 3vsa = −3(vsa − vso ) vsa + vsb + vsc − 3vsb = −3(vsb − vso )

vsa + vsb + vsc − 3vsc = −3(vsc − vso ).

Replacing the above term in (5.153), we get the following. A∆

= −3 [(vsa − vs0 )(vsa − vs0 ) + (vsb − vs0 )(vsb − vs0 ) + (vsc − vs0 )(vsc − vs0 )] = −3[(vsa − vs0 )2 + (vsb − vs0 )2 + (vsc − vs0 )2 ] −3

=



j=a,b,c

(vs j − vs0 )2

(5.154)

The above equation can also be written as, |A∆ |

= −3 [v2sa + v2sb + v2sc + 3 v2s0 − 2 vs0 (vsa + vsb + vsc )] −3 [v2sa + v2sb + v2sc + 3 v2s0 − 6 v2s0 ] = −3 [v2sa + v2sb + v2sc − 3v2s0 ]  

=

=

−3



j=a,b,c

v2s j − 3v2s0

Calculation of cofactors of A∆ We need to calculate ac31 , ac32 , and ac33 . These are computed as following. ac31

= [(vsb − vso ) + β (vsc − vsa ) − (vsa − vso ) − β (vsb − vsc )] = [(vsb − vsa ) − β (vsa + vsb − 2 vsc )]

= [(vsb − vsa ) − β (vsa + vsb + vsc − 3 vsc )] = [(vsb − vsa ) − β (3 vs0 − 3 vsc )] =

− [vsab − 3 β (vsc − vs0 )]

Similarly, ac32

= = = = =

−[(vsb − vso ) + β (vsc − vsa ) − (vsc − vso ) + β (vsa − vsb )]

−[(vsb − vsc ) + β (vsc + vsb − 2 vsa )] −[(vsb − vsc ) + β (vsa + vsb + vsc − 3 vsa )] −[(vsb − vsc ) + β (3 vs0 − 3 vsa )] − [vsbc − 3β (vsa − vs0 )]

(5.155)

256

Power Quality in Power Distribution Systems: Concepts and Applications

and ac33

= [(vsa − vso ) + β (vsb − vsc ) − (vsc − vso ) + β (vsa − vsb )]

= [(vsa − vsc ) − β (vsa + vsc − 2 vsb )] = [(vsa − vsc ) − β (vsa + vsb + vsc − 3 vsb )] = [(vsa − vsc ) − β (3 vs0 − 3 vsb )] = − [vsca − 3β (vsb − vs0 )]

Therefore, the solution of the equation is given by,   ilab − i∗f ab  ilbc − i∗f bc  = ilca − i∗f ca

 a 1  c11 ac21 A∆ ac31  a 1  c11 ac12 A∆ ac13

=

ac12 ac22 ac32 ac21 ac22 ac23

T   ac13 0 ac23   0  ac33 Plavg   ac31 0 ac32   0  ac33 Plavg

(5.156)

From the above equation and substituting the values of cofactors obtained above, we get the following. ilab − i∗f ab

= =

=

ac31 Plavg |A∆ | − [vsab − 3β (vsc − vs0 )]   Plavg −3 ∑ j=a,b,c v2s j − 3 v2s0 [vsab /3 − β (vsc − vs0 )] Plavg ∑ j=a,b,c v2s j − 3 v2s0

From the above equation, the reference compensator current (i∗f ab ) can be given as follows. i∗f ab = ilab −

vsab /3 − β (vsc − vs0 ) Plavg ∑ j=a,b,c v2s j − 3 v2s0

(5.157)

i∗f bc = ilbc −

vsbc /3 − β (vsa − vs0 ) Plavg ∑ j=a,b,c v2s j − 3 v2s0

(5.158)

i∗f ca = ilca −

vsca /3 − β (vsb − vs0 ) Plavg ∑ j=a,b,c v2s j − 3 v2s0

(5.159)

Similarly,

and

257

Control Theories for Load Compensation

When the source power factor is unity, β = 0 and for balanced source voltages (fundamental), vs0 = 0. Substituting these values in the above equations, we get, vsab Plavg 3 ∑ j=a,b,c v2s j vsbc = ilbc − Plavg 3 ∑ j=a,b,c v2s j vsca = ilca − Plavg 3 ∑ j=a,b,c v2s j

i∗f ab

= ilab −

i∗f bc i∗f ca

(5.160)

Earlier, it was established that, v2sab + v2sbc + v2sca = 3





j=a,b,c

v2s j − 3 v2s0



(5.161)

  Replacing 3 ∑ j=a,b,c v2s j − 3 v2s0 in (5.157), (5.158) and (5.159), we get the following. i∗f ab i∗f bc i∗f ca

vsab − 3β (vsc − vs0 ) Plavg (v2sab + v2sbc + v2sca ) vsbc − 3β (vsa − vs0 ) = ilbc − 2 Plavg (vsab + v2sbc + v2sca ) vsca − 3β (vsb − vs0 ) = ilca − 2 Plavg (vsab + v2sbc + v2sca ) = ilab −

(5.162)

For unity power factor and balanced source voltages (fundamental), the reference compensator currents are given as follows.

5.4.3

i∗f ab

=

i∗f bc

=

i∗f ca

=

vsab Plavg 9V 2 vsbc ilbc − Plavg 9V 2 vsca ilca − Plavg 9V 2 ilab −

(5.163)

COMPENSATION FOR THREE-PHASE THREE-WIRE SYSTEM SUPPLYING A PASSIVE LOAD

If we simplify the above for passive load (resistance, inductance and capacitance), with balanced three-phase supply voltage and current phasors at the fundamental

258

Power Quality in Power Distribution Systems: Concepts and Applications

frequency, we can find out the nature of source currents and the equivalent impedance seen by the source. This is illustrated below. I sab

=

I sbc

=

I sca

=

V sab Plavg 9V 2 V sbc Plavg I lbc − I f bc = 9V 2 V sca Plavg I lca − I f ca = 9V 2 I lab − I f ab =

(5.164)

From (5.164) we can write, I sab

= = = =

V sab Plavg V sab ∗ ∗ ∗ = [Real(V sab I lab +V sbc I lbc +V sca I lca )] 9V 2 9V 2  V sab  ∗ ∗ ∗ ab ∗ bc ∗ ca ∗ Real[V V (Y ) +V V (Y ) +V V (Y ) ] sca sca l sab sab l sbc sbc l 9V 2 V sab ab bc bc ca ca 3V 2 Real[(Gab l − jBl ) + (Gl − jBl ) + (Gl − jBl )] 9V 2 V sab ab ca (Gl + Gbc l + Gl ) 3

Similarly, I sbc

=

I sca

=

V sbc ab ca (Gl + Gbc l + Gl ) 3 V sca ab ca (Gl + Gbc l + Gl ) 3

(5.165)

The above equation also implies that, I sa I sb I sc

bc ca = V sa (Gab l + Gl + Gl ) bc ca = V sb (Gab l + Gl + Gl )

=

(5.166)

bc ca V sc (Gab l + Gl + Gl )

From (5.165), it is clear that the three-phase source sees a balanced delta connected load with the resistance values, Rab = Rbc = Rca = G +G3 +G , as shown in Fig. lab lbc lca 5.12(a). This represents an equivalent star-connected load with the resistance values of Ra = Rb = Rc = G +G1 +G as given in (5.166) and illustrated in Fig. 5.12(b). lab lbc lca This ensures that the compensated source currents are balanced and in phase with the respective phase voltage. 5.4.4

EQUIVALENCE BETWEEN THE PASSIVE AND ACTIVE LOAD COMPENSATION FOR THREE-PHASE THREE-WIRE SYSTEM

At this point, a query may arise in the reader’s mind, what is the equivalence between the passive load compensation for a three-phase three-wire system (discussed

Control Theories for Load Compensation

259

Figure 5.12 Three-phase three-wire compensated system (a) delta equivalent and (b) star equivalent

in Section 4.5 of the previous chapter) and the active load compensation scheme based on symmetrical component theory (discussed above)? While the end results are the same, i.e., the compensated source currents are balanced and in phase with the source voltages, what is their impact on the compensator rating and the losses in the system? Which scheme should be called better? These points need investigation. The two compensation schemes are shown in Fig. 5.13. Fig. 5.13(a) shows passive compensation using a purely reactive network as discussed in Section 4.6 and Fig. 5.13(b) illustrates the compensation scheme using an active filter using symmetrical component theory, discussed in the above section. These schemes thus can be referred to as passive and active compensation for three-phase three-wire system. The three-phase delta-connected load consists of unbalanced passive load, i.e., resistance, inductance, and capacitance. The three-phase supply voltages are balanced and sinusoidal.

Figure 5.13 Compensation schemes (a) passive compensation and (b) active compensation

To compare the currents for the compensator, we shall denote if ab = −i f ab , if bc = −i f bc , if ca = −i f ca . Since only fundamental frequency components are present, the currents can be expressed in phasor form to simplify the analysis. Using passive

260

Power Quality in Power Distribution Systems: Concepts and Applications

compensation, the compensated source currents are given as follows. I sa

passive (t)

=

I sb

passive (t)

=

I sc

passive (t)

=

(I lab + I γab ) − (I lca + I γca ) = (I lab − I lca ) + (I γab − I γca ) (I lbc + I γbc ) − (I lab + I γab ) = (I lbc − I lab ) + (I γbc − I γab )

(I lca + I γca ) − (I lbc + I γbc ) = (I lca − I lbc ) + (I γca − I γbc ) (5.167)

Similarly for the active compensation scheme, I sa active (t) I sb active (t) I sc active (t)

























= (I lab + I f ab ) − (I lca + I f ca ) = (I lab − I lca ) + (I f ab − I f ca ) = (I lbc + I f bc ) − (I lab + I f ab ) = (I lbc − I lab ) + (I f bc − I f ab ) = (I lca + I f ca ) − (I lbc + I f bc ) = (I lca − I lbc ) + (I f ca − I f bc )

(5.168)

In the previous section, we have proved that the source currents after compensation, using active filter are balanced and sinusoidal, i.e., I sa = I sa active , I sb = I sb active , and I sc = I sc active , through (5.165), illustrated in Fig. 5.12. Let us verify the nature of source current for load compensation using purely reactive elements (passive compensation) in (5.167). First consider the expression of I sa passive . I sa

passive (t)

=

(I lab + I γab ) − (I lca + I γca ) = (I lab − I lca ) + (I γab − I γca )

= V ab (Ylab +Yγab ) −V ca (Ylca +Yγca )

(5.169)

In the above expression, V ab V bc V ca Ylab

Ylbc Ylca

Yγab Yγbc Yγca

√ √ 3V √ 3V ∠30◦ = ( 3 + j) 2 √ √ = 3V ∠ − 90◦ = − j 3V √ √ 3V √ ◦ (− 3 + j) = 3V ∠150 = 2 ab = Gab l + jBl =

bc = Gbc l + jBl

ca = Gca l + jBl 

=

jBab γ

=

jBbc γ

=

jBca γ

=j

=j =j

 

−Bab l +

Gca − Gbc l √ l 3

−Bbc l +

Gab − Gca l √ l 3

−Bca l +

Gbc − Gab l √ l 3



 

261

Control Theories for Load Compensation

Substituting the above values of V ab , V bc , V ca , Ylab , Ylbc , Ylca , Yγab , Yγbc , and Yγca in (5.169) simplifies to, I sa

passive

bc ca = V sa (Gab l + Gl + Gl )

I sb

passive

bc ca = V sb (Gab l + Gl + Gl )

passive

bc ca V sc (Gab l + Gl + Gl )

I sc

=

(5.170)

Thus, the final compensated source currents using passive compensation, i.e., I sa passive , I sb passive , and I sc passive given in (5.170) are same as the compensated source currents using active compensation, i.e., I sa active , I sb active , and I sc active , given in (5.166). Since the final compensated currents are same, does it mean that, the line    compensator currents are also same for active (I f a , I f b , and I f c ) and passive (I γa , I γb , and I γc ) compensation? Let us find out. Comparing (5.167) and (5.168), we can write, 









I f ab − I f ca

= I γab − I γca =⇒ I f a = I γa 

I f bc − I f ab

= I γbc − I γab =⇒ I f b = I γb

  I f ca − I f bc

= I γca − I γbc =⇒

 I fc

(5.171)

= I γc

The above equation (5.171) indicates that the net compensator currents, as seen from star connected voltage supply, are the same for active and passive compensation. Let    us find out I f ab , I γab , I f bc , I γbc , and I f ca , I γca and compare their values. Using (5.164),  the current, I f ab is given by, 

I f ab = I sab − I lab

(5.172)

ab bc ca ab ab ab From (5.165), I sab = V sab 3 (Gl + Gl + Gl ) and I lab = V sab Yl = V sab (Gl + jBl ).  Thus from (5.172), I f ab is simplified as following.   bc ca Gab  ab ab l + Gl + Gl I f ab = V sab − Gl − j Bl 3   bc ca −2Gab ab l + Gl + Gl (5.173) − j Bl = V sab 3

Similarly,  I f bc

= V sbc



bc ca Gab l − 2Gl + Gl − j Bbc l 3



= V sbc



bc ca Gab l + Gl − 2Gl − j Bca l 3



and,  I f ca

262

Power Quality in Power Distribution Systems: Concepts and Applications

Now we compute, I γab , I γbc , and I γca , as given below.  I γab = V sab Yγab

= V sab jBab γ

− jBab l +

Gca − Gbc j l √ l 3



= V sbc



− jBbc l +

Gab − Gca j l √ l 3



= V sca



− jBca l +

Gbc − Gab j l √ l 3



= V sab

(5.174)

Similarly, I γbc = V sbc Yγbc

= V sbc jBbc γ

I γca = V sca Yγca

= V sca jBca γ

and,

From (5.173) and (5.174), it is clear that, 

I f ab = I γab 

I f bc = I γbc  I f ca

(5.175)

= I γca

Nevertheless from (5.171), 





I γab − I f ab = I γbc − I f bc = I γca − I f ca = K (say)

(5.176) 





Then the relationship between the active phase filter currents (I f ab , I f bc , I f ca ) and passive filter phase currents (I γab , I γbc , I γca ) is given as, 

I f ab = I γab − K 

I f bc = I γbc − K  I f ca

(5.177)

= I γca − K

From (5.173) and (5.174), it can be found that,   bc ca √ Gab − Gca Gab l √ l l − 2Gl + Gl K = 3V +j 3 3

(5.178)

From load currents, I lab , I lbc , I lca and passive compensator currents, I γab , I γbc , I γca , it is found that, I lab + I lbc + I lca I γab + I γbc + I γca + = I lab0 + I γab0 = I sab0 = K 3 3

(5.179)

From (5.177) and (5.179), 





I lab + I lbc + I lca I f ab + I f bc + I f ca + =0 3 3

(5.180)

Control Theories for Load Compensation

263

In (5.179), the terms, I lab0 , I γab0 are load and compensator circulating currents. Their sum, i.e., I sab0 = K, is the net circulating current in the compensated delta network. This actually means that in the passive compensation scheme, the net circulating current exists in the compensated system. While from (5.180), it is evident that there is no circulating current in the active compensation scheme. This is the main difference between the active and passive compensation schemes for three-phase three-wire systems, although the source currents after compensation are the same for both schemes. For the requirement of zero circulating current after compensation, the active compensation scheme should be chosen.

Figure 5.14 Compensation of delta connected load (a) without compensation, (b) active compensation, and (c) passive compensation

These situations are illustrated through circuits shown in Fig. 5.14(a)–(c). Fig. 5.14(a) shows a delta-connected load without compensation. Since the load is

264

Power Quality in Power Distribution Systems: Concepts and Applications

unbalanced, the phase currents, ilab , ilbc , ilca have positive, negative, and zero sequence currents, as expressed below. ilab

= ilab+ + ilab− + ilab0

ilbc

= ilbc+ + ilbc− + ilbc0

ilca

= ilca+ + ilca− + ilca0

(5.181)

The zero sequence current, ilab0 = ilbc0 = ilca0 , that flows in the delta structure, is also known as circulating current, as shown in Fig. 5.14(a). In the active compensation scheme shown in Fig. 5.14(b), the compensator generates an equal circulating current in the opposite direction, thus leading to zero circulating current in the delta. The active compensator also compensates negative sequence currents for each phase. The resultant phase current components are the summation of positive sequence components of load and the compensator, which makes line currents balanced with the unity power factor. The source sees the compensated load as it is, i.e., balanced resistive load. In the case of the passive load compensation scheme, as illustrated in 5.14(c), the net circulating current is not zero. In fact, the load circulating current and passive compensator circulating current are not related and these will add up to some non-zero finite value. However, negative sequence components of the load currents are canceled by the compensator negative sequence currents. Due to this, the actual compensated load consists of three different impedance elements, however, when looked at from the source, it is seen as a balanced resistive load, similar to that of active compensation. Example 5.6. Consider a three-phase balanced system shown in Fig. 5.11. The supply voltages are: V a = 230∠0◦ , V b = 230∠−120◦ , V c = 230∠120◦ . The impedances of delta connected load are; Zab = 5 + j 12 Ω, Zbc = 3 + j 4 Ω and Zca = 12 − j 13 Ω. Using instantaneous symmetrical component theory, (a) Find the line currents (ila , ilb , ilc ) for the above delta connected load. (b) Find reference filter currents (i f ab , i f bc , i f ca ) of an ideal compensator such that the source currents are balanced sinusoids and in phase with their respective phase voltages. Also, find out the time expressions for the source currents and check whether these are balanced and in phase with their respective voltages. (c) Based on these reference filter currents, identify the nature and values of compensator emulated impedances, i.e., Z f ab , Z f bc , Z f ca . Compare their values with those obtained using the theory load compensation technique discussed in Section 4.5. (d) Compute the active and reactive powers of the load, compensator, and source. Verify that, Ps = Pl , Qs = 0, Pf = 0, and Q f = Ql . (e) Compare the above results with those of passive compensation (purely reactive network) for the system.

Control Theories for Load Compensation

265

Solution: (a) To compute the source currents, first line to line currents are calculated. These are given below. I lab

=

I lbc

=

I lca

=

V sa −V sb V sab 398.37∠30◦ = 30.64∠ − 37.38◦ A = = Zlab Zlab 5 + j 12 V sb −V sc V sbc 398.37∠ − 90◦ = 79.67∠ − 143.13◦ A = = Zlbc Zlbc 3+ j4 V sc −V sa V sca 398.37∠150◦ = 22.52∠ − 162.71◦ A = = Zlca Zlca 12 − j 13

By applying KCL at load points, the line currents are calculated as below. I la

=

I lb

=

I lc

=

I lab − I lca = 47.37∠ − 14.56◦ A

I lbc − I lab = 92.80∠ − 161.66◦ A I lca − I lbc = 58.94∠44.22◦ A

Therefore, the time expressions for the line currents are given as follows. √ ila (t) = 47.37 2 sin(ω t − 14.56◦ ) A √ ilb (t) = 92.80 2 sin(ω t − 161.66◦ ) A √ ilc (t) = 58.94 2 sin(ω t + 44.22◦ ) A It can be seen that the source currents are not balanced in magnitude as well as in phase angles and hence cause various reactive powers and power factors in different phases. (b) Now to obtain balanced source currents at unity power factor, the filter currents (i f ab , i f bc , i f ca ) are computed using (5.163), as given below. i∗f ab i∗f bc i∗f ca

vsab Plavg 9V 2 vsbc = ilbc − Plavg 9V 2 vsca = ilca − Plavg 9V 2 = ilab −

Since only fundamental frequency components are involved in computation, the above equation can be used in the phasor domain. Since the line currents and line to line voltages are known, the remaining term, Plavg is calculated as follows.  ∗ ∗ ∗  Plavg = Real V sab I lab +V sbc I lbc +V sca I lca = 29824 W

266

Power Quality in Power Distribution Systems: Concepts and Applications

Finally, reference filter currents are computed as follows. I f ab

= I f bc

= =

I f ca

V sab 398.37∠30◦ × 29824 Plavg = 30.64∠ − 30.38◦ − 2 9V 9 × 2302 ◦ 31.2∠−84.96 A 398.37∠ − 90◦ × 29824 V sbc ◦ I lbc − P = 79.67∠ − 143.13 − lavg 9V 2 9 × 2302 ◦ 67.71∠−160.28 A V sca 398.37∠150◦ × 29824 ◦ I lca − P = 22.52∠ − 162.71 − lavg 9V 2 9 × 2302 ◦ 19.17∠−89.67 A

= I lab −

= =

Therefore, the time expressions for the compensator currents are given as follows. √ i f ab (t) = 31.2 2 sin(ω t − 84.96◦ ) A √ i f bc (t) = 67.71 2 sin(ω t − 160.28◦ ) A √ i f ca (t) = 19.17 2 sin(ω t − 89.67◦ ) A After compensation, the source currents are computed as follows. I sa I sb I sc

= (I lab − I f ab ) − (I lca − I f ca ) = 43.22∠0◦ A

= (I lbc − I f bc ) − (I lab − I f ab ) = 43.22∠ − 120◦ A = (I lca − I f ca ) − (I lbc − I f bc ) = 43.22∠120◦ A

Therefore, the time expressions of source currents after compensation are given as follows. √ isa (t) = 43.22 2 sin(ω t) A √ isb (t) = 43.22 2 sin(ω t − 120◦ ) A √ isc (t) = 43.22 2 sin(ω t + 120◦ ) A Thus, it is seen that the source currents after compensation becomes balanced and are in phase with their respective source voltages. (c) Based on the line to line voltages and the filter currents, the compensator impedances Z f ab , Z f bc , Z f ca are computed as follows. Z f ab

=

Z f bc

=

Z f ca

=

V sab 398.37∠30◦ = = 5.3885 − j 11.5748 Ω −31.2∠ − 84.96◦ −I f ab

V sbc 398.37∠ − 90◦ = = −1.9854 − j 5.5382 Ω −67.71∠ − 160.28◦ −I f bc 398.37∠150◦ V sca = = 10.495 + j 17.9359 Ω −19.17∠ − 89.67◦ −I f ca

(5.182)

267

Control Theories for Load Compensation

As seen from the expressions, the compensator impedance Z f ab is a combination of a resistance and a capacitive reactance, Z f bc has a negative resistance and a capacitive reactance, and Z f ca has a resistance and an inductive reactance. This negative resistance is realized using active power flow between the phases of the compensator. Since this is a three-phase three-wire system, the three-phase unbalanced delta load can also be compensated using purely reactive components, as discussed in Section 4.6. Based on this, the compensator admittances are computed as follows.   Gca − Gbc ab ab ab l √ l = j 0.0238  Yγ = jBγ = j −Bl + 3   Gab − Gca bc bc bc l √ l Yγ = j 0.1549  (5.183) = jBγ = j −Bl + 3   Gbc − Gab ca ca ca l √ l Yγ = j 0.0107  = jBγ = j −Bl + 3 Thus, the compensator impedances are as follows. Zγab

=

Zγbc

=

Zγca

=

1 = − j 41.91 Ω Yγab 1 = − j 6.45 Ω Yγbc 1 = − j 93.7 Ω Yγca

(5.184)

As it can be seen from (5.182) and (5.184), the nature of compensator impedances are not the same. In active filtering, the compensator emulates negative resistance such that the net real power of the active power filter is zero, nevertheless, real power flows in the phases of the active filter. In the case of the passive filter, the passive compensator impedances are purely reactive. However, in both cases, the compensated source currents are the same, i.e., balanced and in phase with their phase voltages. (d) The active and reactive powers of the load, compensator, and source can be found as given below.  ∗ ∗ ∗  Pl = Real V sab I lab +V sbc I lbc +V sc I lca = 29824 W  ∗ ∗ ∗  Ql = Imag V sab I lab +V sbc I lbc +V sc I lca = 30069 VAr Pf Qf

 ∗ ∗ ∗  = Real V sab I f ab +V sbc I f bc +V sca I f ca = 0 W  ∗ ∗ ∗  = Imag V sab I f ab +V sbc I f bc +V sca I f ca = −30069 VAr

268

Power Quality in Power Distribution Systems: Concepts and Applications

Ps Qs

 ∗ ∗ ∗  = Real V sa I sa +V sb I sb +V sc I sc = 29824 W  ∗ ∗ ∗  = Imag V sa I sa +V sb I sb +V sc I sc = 0 VAr

From the above, it is observed that after compensation, Ps = Pl = 29824 W, Qs = 0 VAr, Pf = 0 W, Q f = Ql = 30069 VAr. This ensures that the source only supplies average load power while the compensator supplies reactive power of the load. (e) Passive compensation method Using (5.174), the passive compensator currents are computed as follows:   ca − Gbc G ab l √ l = 9.5∠120◦ A I γab = V sab Yγab = V sab jBab γ = V sab − jBl + j 3   Gab − Gca bc bc bc l √ l = 61.73∠0◦ A I γbc = V sbc Yγ = V sbc jBγ = V sbc − jBl + j 3   bc − Gab G ca l √ l = 4.25∠ − 120◦ A I γca = V sca Yγca = V sca jBca γ = V sca − jBl + j 3 Therefore, the source currents are given as, = (I lab − I γab ) − (I lca − I γca ) = 43.22∠0◦ A

I sa

= (I lbc − I γbc ) − (I lab − I γab ) = 43.22∠ − 120◦ A

I sb

= (I lca − I γca ) − (I lab − I γab ) = 43.22∠120◦ A

I sc

As can be seen from the above equation, the source currents are the same as that in the case of the active compensator. Nevertheless, the individual phase compensator currents are different. As discussed earlier, the difference between the two schemes accounts for the circulating current in the compensated delta circuit. The net circulating current after passive compensation is: I lab + I lbc + I lca I γab + I γbc + I γca + = 22.94∠ − 95.03◦ A 3 3 And the net circulating current for the active compensation method is zero as given in the following. 





I lab + I lbc + I lca I f ab + I f bc + I f ca + =0 3 3 Further, it can be seen that, 





I γab − I f ab = I γbc − I f bc = I γca − I f ca = 22.94∠ − 95.03◦ A

269

Control Theories for Load Compensation

Therefore, the difference between the compensator currents for each phase is the difference between the circulating currents for the two compensation schemes. The passive compensator powers are:  ∗ ∗ ∗  Pγ = Real V sab I γab +V sbc I γbc +V sca I γca = 0 W  ∗ ∗ ∗  Qγ = Imag V sab I γab +V sbc I γbc +V sca I γca = 30069 VAr

Due to the opposite direction of active filter currents to that of passive filter currents, Qγ = −Q f . The phase wise ratings of the active and passive compensators are: Sγab

= Qγab = 3.78 kVA, capacitive

Sγbc

= Qγbc = 24.59 kVA, capacitive

Sγca

= Qγbc = 1.69 kVA, capacitive

and, S f ab

= 12.43 kVA

S f bc

= 26.97 kVA

S f ca

= 7.64 kVA

As evident from these values, the phase wise ratings of the active compensator are more as compared to those of the passive compensator. It is due to the additional circulating current that is compensated by the active compensator. However, this may not always be the case, as it depends upon the nature of the load.

5.5

SUMMARY

This chapter has primarily focused on control theories for the compensation of a three-phase load with unbalanced harmonic components. Two theories, i.e., instantaneous reactive power theory and instantaneous symmetrical component theory have been discussed in detail with illustrations. Some conceptual anomalies of the instantaneous reactive power theory are also discussed. A generic, neutral clamped voltage source converter topology is discussed and its state space model is developed to simulate three-phase compensated systems with active power filters on a digital platform. Various components of the three-phase compensated system, such as the computation of losses in the converter using dc bus voltage regulation, computation of average load power, and other power components, are described. A comparison between active and passive compensation schemes is brought out with mathematical and circuit descriptions for a three-phase three-wire system. A number of examples are presented to explain the concepts, followed by subjective and numerical problems.

270

5.6

Power Quality in Power Distribution Systems: Concepts and Applications

PROBLEMS

P 5.1 Explain the concept of instantaneous reactive power (pq) theory. Based on this concept: (a) Define instantaneous active, p(t) and instantaneous reactive, q(t) powers for a given three-phase four-wire system using αβ 0 transformation. (b) Show that power is constant in both abc and αβ 0 frames of reference. (c) Explain how you will extract reference compensator currents based on this theory. (d) Knowing reference currents, how do you realize these currents using a voltage source inverter? Assume terms and notations for three-phase system quantities and circuits to illustrate the above. P 5.2 With the help of a figure, illustrate the power flows between the source, load, and compensator in αβ 0 frame of reference. P 5.3 Explain how you will extract the compensator reference currents using instantaneous reactive power theory (pq theory) with the help of equations and a schematic diagram. Also, derive the expression for these currents. P 5.4 Develop state space model for the three-phase, four-wire compensated system. The compensator is realized using a neutral clamped voltage source inverter. How this state space model is solved using MATLAB program? P 5.5 Explain how the dc link voltage supporting the voltage source inverter operation is regulated to reference value. P 5.6 Discuss how the extracted reference compensator currents are synthesized using a neutral clamped voltage source inverter with a hysteresis current control scheme. P 5.7 Describe the instantaneous symmetrical component theory to compensate for an unbalanced nonlinear load for the star as well as delta connected load. P 5.8 Show that the locus of real versus imaginary part of instantaneous positive (or negative) sequence symmetrical components of balanced voltages at the fundamental frequency, is a circle. Discuss the applications of this in identifying the nature of voltage and current waveforms. P 5.9 Consider a three-phase four-wire compensated system in the figure below. The three-phase voltages and load impedances are as follows. √ vsa (t) = 230 2 sin ω t V √ vsb (t) = 230 2 sin(ω t − 120◦ ) V √ vsc (t) = 230 2 sin(ω t + 120◦ ) V

Control Theories for Load Compensation

271

Figure 5.15 Related to problem P 5.9

Load impedances: Zla = 3 + j 4 Ω, Zlb = 5 − j 12 Ω, Zlc = 7 + j 9 Ω. Using instantaneous symmetrical component theory, (a) Find reference filter currents (i f a , i f b , i f c ) of an ideal compensator such that source currents are balanced sinusoids and in phase with their respective phase voltages. Also, find the active and reactive powers of the compensator. (b) Based on these reference filter currents, identify the nature and values of Z f a , Z f b , Z f c . Are these practically realizable using passive impedances? (c) If you realize the compensator using a voltage source inverter, explain for the actual (not ideal) compensator how expressions of reference filter currents are modified to maintain dc link voltage supported by dc storage capacitors to a constant reference value. (d) While generating reference currents, which parameters play significant roles in deciding the dynamic performance of the compensator? P 5.10 Refer to the above question to answer the following using instantaneous reactive power (pq) theory and verify answers. (a) α, β and 0 components of three-phase supply voltages (vsα , vsβ , vs0 ,) and load currents (ilα , ilβ , il0 ). (b) The various components of instantaneous and average powers i.e., pα , pβ , p0 , qαβ . (c) Verify these powers on the basis of their average values, i.e., Pabc = Pα + Pβ + P0 = Pαβ 0 and Qabc = −Qαβ . (d) Reference compensator currents in αβ 0 frame, i.e., (i f α , i f β , i f 0 ) and abc frame, i.e., (i f a , i f b , i f c ) so that source supplies balanced currents from the source with unity power factor relationship with their voltages. (e) Check whether the source currents are balanced and sinusoidal and in phase with their voltages. (f) Check whether ps (t) = Pl , qs (t) = 0, p f (t) = 0 and q f (t) = ql ?

272

Power Quality in Power Distribution Systems: Concepts and Applications

P 5.11 Consider a three-phase four-wire compensated system. The three-phase source voltages and load impedances are as follows. √ vsa (t) = 230 2 sin ω t V √ vsb (t) = 230 2 sin(ω t − 2π /3) V √ vsc (t) = 230 2 sin(ω t + 2π /3) V Load impedances: Zla = 6 + j 8 Ω, Zlb = 3 − j 4 Ω, Zlc = 12 + j 5 Ω. (a) Resolve system voltages and load currents into α, β and 0 components. (b) Based on α, β , and 0 components of voltages and currents, compute the following. (i) Instantaneous α-axis (pα ), β -axis (pβ ), pαβ , 0-axis (p0 ) instantaneous powers. (ii) Instantaneous reactive power, qαβ (t). (iii) Then, compute real power (Pl ) and reactive power (Ql ) of the load. (c) If an active power filter is employed, what are compensator currents in αβ 0 and abc frames of reference? (d) Compute the source currents (isa , isb , isc ), after compensation. P 5.12 Verify answers to parts (c) and (d) of the above problem, using instantaneous symmetrical component theory. P 5.13 Consider a three-phase three-wire compensated system. The three-phase source voltages are balanced with a phase voltage of 230 V rms, 50 Hz and supplies three-phase unbalanced delta load with Zlab = 10 + j 8 Ω, Zlbc = 12 + j 5 Ω, Zlca = 12 − j 8 Ω. (a) Draw the schematic of the above system. (b) Compute reference compensator phase currents, i f ab , i f bc , i f ca , using instantaneous symmetrical component theory, considering an ideal compensator such that source currents are balanced sinusoid and in phase with their respective phase voltages. (c) Compute source currents after compensation. (d) Find active and reactive powers of the compensator and also source active and reactive, and load active and reactive powers. Establish a relationship between these powers. (e) Compare it with a compensation scheme with a purely reactive network. P 5.14 Consider the delta-star connected ideal transformer with 1:1 turn ratio shown in Fig. 5.16. On the secondary side, a load of R ohms is connected between the phase-a and neutral n.

Control Theories for Load Compensation

273

Figure 5.16 Related to problem P 5.14

Assuming balanced source voltages, compute the following. (a) Time domain expressions of currents in each phase on both primary and secondary sides. (b) Does the load require reactive power from the source? If any, find its value. Also, compute the reactive power on each phase of either side of the transformer. (c) Also, determine active and reactive powers on each phase and overall active power on either side of the transformer. (d) If you have a similar arrangement with a balanced load and the same output power, comment upon the rating of line conductors, and transformers. P 5.15 In the above question, the primary and secondaries are interchanged and the load is connected between phase-a and b, as shown in Fig. 5.9. Answer points (a)–(d) in the question. P 5.16 Consider a three-phase balanced system shown in Fig. 5.11. The supply voltages are: V a = 230∠0◦ , V b = 230∠ − 120◦ , V c = 230∠120◦ . The impedances of delta connected load are, Zab = 3 + j 4 Ω, Zbc = 5 + j 10 Ω and Zca = 5 − j 13 Ω. Using instantaneous symmetrical component theory, (a) Find the line currents (ila , ilb , ilc ) for the above delta connected load. (b) Find reference filter currents (i f ab , i f bc , i f ca ) of an ideal compensator such that source currents are balanced sinusoids and in phase with their respective phase voltages. Also, find out the time expressions of the source currents after compensation. Check whether these are balanced and in phase with their respective voltages. (c) Based on these reference filter currents, identify the nature and values of Z f ab , Z f bc , Z f ca . Verify these values obtained using the theory of load compensation technique discussed in Section 4.5 and comment upon the values.

274

Power Quality in Power Distribution Systems: Concepts and Applications

(d) Compute the active and reactive powers of the load, compensator, and source. Verify that, Ps = Pl , Qs = 0, Pf = 0, and Q f = Ql . (e) Compare the above results with those of passive compensation (purely reactive network) for the system. P 5.17 Consider a three-phase four-wire compensated system as shown in Fig. 5.15. The three-phase source voltages and load impedances are as follows. √ √ ◦ Supply voltages: vsa = 220 2 sin √ ωt, vsb = 150 ◦ 2 sin(ωt − 150 ) vsc = 300 2 sin(ωt + 100 ) V Load impedances: Zla = 12 + j 5 Ω, Zlb = 3 − j 4 Ω, Zlc = 25 − j 7 Ω. (a) Using instantaneous symmetrical component theory, find reference filter currents (i∗f a , i∗f b , i∗f c ) in time domain considering an ideal compensator such that source currents are balanced sinusoids and in phase with their respective positive sequence fundamental phase voltages, while source supplies the same load power. (b) Find active (Pf ) and reactive (Q f ) powers of the compensator and also source active (Ps ) and reactive (Qs ) and load active (Pl ) and reactive (Ql ) powers. Check whether, Pf = 0 and Q f = −Ql and Ps = Pl , Qs = 0. [Hint: Since the three-phase voltages are unbalanced at the fundamental frequency, the compensated source currents cannot be balanced to meet the condition, vsa isa + vsb isb + vsc isc = Plavg (a constant value). To meet this condition with balanced fundamental source currents, fundamental positive sequence voltages should be computed to be used in the algorithm to find reference filter currents].

P 5.18 Consider a three-phase four-wire compensated system as shown (refer Fig. 5.15). Due to some abnormal condition, phase-c supply voltage becomes zero. Thus, the three-phase unbalanced voltages are expressed as follows. √ √ Supply voltages: vsa = 220 2 sin ωt, vsb = 220 2 sin(ωt − 150◦ ), and vsc = 0 V. The load impedances are: Zla = 9 + j3, Zlb = 6 − j8, Zlc = 5 + j12 Ω. (a) Find reference filter currents (i f a , i f b , i f c ) of an ideal compensator such that source currents ( isa , isb , isc ) are balanced sinusoids and in phase with their respective fundamental positive sequence phase voltages, while source supplies the same load power. Also, compute compensated source currents. (b) With the above compensation scheme, compute load, compensator, and source active (P) and reactive (Q) powers comment upon the values of these powers. Will the instantaneous power from the supply be constant (i.e., Plavg )?

Control Theories for Load Compensation

275

REFERENCES 1. N. Mohan, H. A. Peterson, G. R. D. W. F. Long, and J. J. Vithayathil, “Active filters for ac harmonic suppression,” IEEE/PES Winter Meeting, New York City, USA, pp. 529–535, Jan./Feb. 1977. 2. L. Gyugyi and E. C. Strycula, “Active AC Power Filters,” IEEE-IAS Annual Meeting Record, pp. 529–535, 1976. 3. S. Bhattacharya and D. Divan, “Active filter for utility interface of industrial load,” IEEE Transactions on Industry Application, vol. 32, no. 6, pp. 1312–1322, Nov./Dec. 1996. 4. S. Bhattacharya and D. Divan, “Active filter for utility interface of industrial load,” Proceedings IEEE Conf. PEDES ’98, New Delhi, pp. 1079–1084, Jan. 1998. 5. H. Sasaki and T. Machida, “A new method to eliminate ac harmonic currents by magnetic flux compensation considerations on basic design,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-90, no. 5, pp. 2009–2019, 1971. 6. H. Akagi, Y. Kanazawa, and A. Nabae, “Instantaneous reactive power compensators comprising switching devices without energy storage components,” IEEE Transactions on Industry Applications, vol. IA-20, no. 3, pp. 625–630, 1984. 7. A. Ghosh and A. Joshi, “A new approach to load balancing and power factor correction in power distribution system,” IEEE Transactions on Power Delivery, vol. 15, no. 1, pp. 417–422, Jan. 2000. 8. A. Ghosh and A. Joshi, “The use of instantaneous symmetrical components for balancing a delta connected load and power factor correction,” Electric Power System Research, vol. 54, no. 1, pp. 67–74, April 2000. 9. R. A. Otto, T. H. Putman, and L. Gyugyi, “Principles and applications of static, thyristor-controlled shunt compensators,” IEEE Transactions on Power Apparatus and Systems, vol. PAS-97, no. 5, pp. 1935–1945, 1978. 10. L. Gyugyi, “Reactive power generation and control by thyristor circuits,” IEEE Transactions on Industry Applications, vol. IA-15, no. 5, pp. 521–532, 1979. 11. A. Ghosh and A. Joshi, “A new approach to load balancing and power factor correction in power distribution system,” IEEE Transactions on Power Delivery, vol. 15, no. 1, pp. 417–422, 2000. 12. A. Ghosh and A. Joshi, “A new method for load balancing and power factor correction using instantaneous symmetrical components,” IEEE Power Engineering Review, vol. 18, no. 9, pp. 60–61, 1998. 13. A. Ghosh and A. Joshi, “The use of instantaneous symmetrical components for balancing a delta connected load and power factor correction,” Electric Power Systems Research, vol. 54, no. 1, pp. 67–74, 2000.

276

Power Quality in Power Distribution Systems: Concepts and Applications

14. A. Ghosh and G. Ledwich, Power Quality Enhancement Using Custom Power Devices, ser. Power Electronics and Power Systems. Springer US, 2012. 15. E. Watanabe, R. Stephan, and M. Aredes, “New concepts of instantaneous active and reactive powers in electrical systems with generic loads,” IEEE Transactions on Power Delivery, vol. 8, no. 2, pp. 697–703, 1993. 16. M. Aredes and E. Watanabe, “New control algorithms for series and shunt threephase four-wire active power filters,” IEEE Transactions on Power Delivery, vol. 10, no. 3, pp. 1649–1656, 1995. 17. T. Furuhashi, S. Okuma, and Y. Uchikawa, “A study on the theory of instantaneous reactive power,” IEEE Transactions on Industrial Electronics, vol. 37, no. 1, pp. 86–90, 1990. 18. M. Aredes, J. Hafner, and K. Heumann, “Three-phase four-wire shunt active filter control strategies,” IEEE Transactions on Power Electronics, vol. 12, no. 2, pp. 311–318, 1997. 19. F. Peng, H. Akagi, and A. Nabae, “A new approach to harmonic compensation in power systems-a combined system of shunt passive and series active filters,” IEEE Transactions on Industry Applications, vol. 26, no. 6, pp. 983–990, 1990. 20. H. Akagi, “Trends in active power line conditioners,” IEEE Transactions on Power Electronics, vol. 9, no. 3, pp. 263–268, 1994. 21. H. Fujita and H. Akagi, “A practical approach to harmonic compensation in power systems-series connection of passive and active filters,” IEEE Transactions on Industry Applications, vol. 27, no. 6, pp. 1020–1025, 1991. 22. Mahesh K. Mishra, A. Ghosh, and A. Joshi, “A new STATCOM topology to compensate loads containing ac and dc components,” in IEEE Power Engineering Society Winter Meeting. Conference Proceedings, vol. 4, 2000, pp. 2636– 2641. 23. V. George and Mahesh K. Mishra, “DSTATCOM topologies for three phase high power applications,” International Journal of Power Electronics, vol. 2, no. 2, pp. 107–124, 2010. 24. I. J. Nagrath and M. Gopal, Control Systems Engineering. New Age International Pvt., New Delhi, 2005. 25. Mahesh K. Mishra and K. Karthikeyan, “An investigation on design and switching dynamics of a voltage source inverter to compensate unbalanced and nonlinear loads,” IEEE Transactions on Industrial Electronics, vol. 56, no. 8, pp. 2802–2810, 2009. 26. L. Czarnecki, “On some misinterpretations of the instantaneous reactive power pq theory,” IEEE Transactions on Power Electronics, vol. 19, no. 3, pp. 828–836, 2004.

Compensation 6 Voltage Using Dynamic Voltage Restorer

6.1

INTRODUCTION

Power system should ensure good quality of electric power supply, which means voltage and current waveforms should be balanced and sinusoidal. Furthermore, the voltage levels on the system should be within reasonable limits, such as ±10% of their rated value and the frequency of grid supply should be normally within limits of ±1% of the nominal value, depending upon the grid codes of the country [1]– [5]. If the voltage and frequency are more or less than their pre-specified value, the performance of the equipment is compromised. In case of low voltages, picture on television starts rolling, the torque of induction motor reduces to the square of voltage and therefore there is a need for voltage compensation. The frequency level control is achieved through governor dynamics of the synchronous generators in the power system. However, grid frequency control is not part of discussion in this book. Voltage related disturbances and variations are more common in power distribution network, which arise due to various reasons such as voltage drops in the power network, different types of faults in the network, energization of large loads and capacitors bank, switching off heavy loads, loose connections in wiring, switching transients, etc. These variations in the voltage are termed and quantified as voltage sag, swell, interruptions, fluctuations, unbalance, waveform distortion, etc. The detailed descriptions of these terms were given in the Chapter 1. One study reveals that approximately 60–70% of disturbances in power system are due to single-line to ground faults, 11% due to line-line faults and 6% due to three-phase symmetrical faults. Due to finite impedance of power lines, distribution feeders and transformers, these faults get translated into voltage sag, swell or interruption at the point of common coupling, where various loads are connected [6]–[9]. To compensate voltage, two broad categories of voltage regulating devices such as passive and active are used. Each of these devices has either series or shunt configuration. In this chapter we shall focus on series active power filter, which is termed as Dynamic Voltage Restorer (DVR). This is one of the variants of voltage source converter based compensating device. Various aspects of DVR such as topology, analysis and design of series active voltage compensator are described in [10]–[14]. In this chapter, we shall mainly focus on voltage variation, specially short term variations, such as voltage sag, swell and their mitigation using a series active power filter, i.e., DVR.

DOI: 10.1201/9781032617305-6

277

278

6.2

Power Quality in Power Distribution Systems: Concepts and Applications

METHODS TO REGULATE VOLTAGE

In order to regulate the load bus voltage within allowable limits, various compensating devices such as listed below can be used. 1. Shunt capacitor 2. Series capacitor 3. Synchronous capacitor 4. Tap changing transformer 5. Booster transformer 6. Static synchronous series capacitor 7. Dynamic voltage restorer The first six methods are normally employed at transmission level while the DVR is used in power distribution network to protect any voltage variation at the load bus connected to the sensitive and critical electrical units. The DVR is a series connected custom power device used to mitigate the voltage unbalance, sags, swells, harmonics and any abrupt changes due to abnormal conditions in the system. In the following section, dynamic voltage restorer will be described in detail.

6.3

DYNAMIC VOLTAGE RESTORER (DVR)

A dynamic voltage restorer is a solid-state inverter based on injection of voltage in series with a power distribution system [15]–[18]. A single-line diagram of a DVR connected to a power distribution system is shown in Fig. 6.1. The dc side of the DVR is connected to an energy source or an energy storage device, while its ac side is connected to the distribution feeder by an interfacing transformer. In the figure, vs (t) represents the supply voltage, vt (t) represents the terminal voltage and vl (t) represents the load voltage. The inductance, Ls in series with resistance, Rs represents the feeder impedance of the line. Since DVR is a series connected device, the source current, is (t) is the same as the load current, il (t). Also, note in the figure, v f (t) is

Figure 6.1 A single-line diagram of DVR compensated system

Voltage Compensation Using Dynamic Voltage Restorer

279

DVR injected voltage in series with the line such that the load voltage is maintained sinusoidal at the nominal value. The three-phase DVR compensated system is shown in Fig. 6.2. It is assumed that the transmission line has the same impedance in all three phases. A DVR unit which is represented in Figs. 6.1 and 6.2, have following components, as shown in Fig. 6.3 [19]– [21].

Figure 6.2 A three-phase DVR compensated system

Figure 6.3 Schematic diagram of a DVR based compensation in a distribution system

1. Voltage source inverter (VSI) 2. Filter capacitors and inductors 3. Injection transformer 4. DC storage system Some other important issues, i.e., how much voltage should be injected in series using the appropriate algorithm, choice of suitable power converter topology to synthesize voltage, and design of filter capacitor and inductor components, have to be addressed while designing the DVR unit.

280

6.4

Power Quality in Power Distribution Systems: Concepts and Applications

OPERATING PRINCIPLE OF DVR

Consider a DVR compensated single-phase system as shown in Fig. 6.1. Let us assume that the source voltage is 1.0 p.u. and we want to regulate the load voltage to 1.0 p.u. Let us denote the phase angle between V s and V l as δ . In this analysis, harmonics are not considered. Further, during DVR operation, minimal real power is required to counter losses in the inverter and the filter components. These losses, for the time being, are considered to be zero. Also, in this analysis, only reactive power exchange is considered for the DVR operation. This condition implies that the phase difference between V f and I s should be 90◦ . Let us first consider a general case to understand the concept. The DVR equivalent circuit with fundamental voltages and current is shown in Fig. 6.1. Applying Kirchhoff’s voltage law in the circuit, V s +V f

= I s (Rs + jXs ) +V l

V s +V f

= I s Zs +V l .

(6.1)

Note that in above circuit I s = I l = I. The load voltage V l can be written in terms of load current and load impedance as given below. V s +V f = I (Zs + Zl )

(6.2)

In the above equation, V l = I Zl = I (Rl + j Xl ). Using (6.1), the source voltage can be expressed as in the following. V s = V l + I Rs − (V f − jI Xs )

(6.3)

In the above equation, the DVR voltage, V f is associated with the term j I Xs to indicate that the V f is along phasor j I Xs . This is required for the DVR operation without

involvement of real power. Substituting I = VZ l in (6.2), the relationship between load, l source, and DVR voltages can be expressed as below.   V s +V f Vl = (6.4) Zl Zs + Zl

The equation indicates that the load voltage can be controlled using appropriate DVR voltage, V f . The following example illustrates the DVR compensation under various conditions of load and feeder impedances. Example 6.1. Let us apply the condition to maintain load voltage same as source voltage, i.e., V l = V s and discuss the feasibility of the DVR voltage in series with the line as shown in Fig. 6.1. Consider the following cases. (a) Line resistance is negligible with Zs = j0.25 p.u. and Zl = 0.5 + j0.25 p.u. (b) When the line resistance is not negligible, i.e., Zs = 0.45 + j0.25 p.u. and Zl = 0.5 + j0.25 p.u.

281

Voltage Compensation Using Dynamic Voltage Restorer

Solution: (a) When line resistance is negligible The above condition implies that, Rs = 0. Without DVR, the load terminal voltage V l can be given as follows. Il =

Vs 1.0∠0◦ = 1.0 − j1.0 = 1.4142∠ − 45◦ p.u. = Zl + Zs 0.5 + j0.5

Therefore the load voltage is given as following. V l = Zl I l = 0.5590∠26.56◦ × 1.4142∠ − 45◦ = 0.7906∠ − 18.43◦ p.u. This is illustrated in Fig. 6.4(a). Thus, the load voltage has reduced by 21%. Now it is desired to maintain the load voltage same as the supply voltage in magnitude and phase angle. Thus, substituting V s = V l in equation (6.1), we get, V s +V f

=

I (Rs + jXs ) +V l

Vf

=

Vf

=

I (Rs + jXs ) jXs V l , since, Rs = 0 Zl



and

I = V l /Zl

Neglecting the resistance part of the feeder impedance, Zs = j0.25, the DVR voltage can be computed as above, i.e., Vf

= =

j0.25 1.0∠0◦ 0.5 + j0.25 0.4472∠63.43◦ p.u.

for V l = 1.0∠0◦

From the above, the line current is computed as follows. Is =

Vl 1.0∠0◦ = = 1.7889∠ − 26.56◦ p.u.. Zl 0.5590∠26.56◦

It is to be noted that, although V s = V l = 1.0∠0◦ p.u., it does not imply that power does not flow from source to the load. In fact the total effective source  voltage is V s = V s + V f = 1.2649∠18.8◦ p.u. Therefore, the effective source voltage is leading the load voltage by an angle of 18.43◦ . This ensures the power flow from the source to the load. This is illustrated by drawing a phasor diagram in Fig. 6.4(b). (b) When line resistance is not negligible For this case, Zl = Rl + j Xl = 0.5 + j 0.25 p.u. Substituting V s = V l in (6.1), we get the following. V f = (Rs + jXs )I = Rs I s + jXs I s .

282

Power Quality in Power Distribution Systems: Concepts and Applications

Figure 6.4 Load voltage (a) without DVR and (b) with DVR

From the above equation, it is indicated that DVR voltage has two components, one is in phase with I s and the other is in phase quadrature with I s . This implies that for a finite value of feeder resistance, it is not possible to maintain V l = V s without active power supplied from the DVR to the load. This is due to the presence of in-phase component of the DVR voltage in the above equation. This is illustrated in the phasor diagram given in Fig. 6.5. It is to be noted that V l = V s is not a necessary and useful condition to regulate the load voltage. If the amplitude of load voltage equals to the supply voltage, i.e., Vl = Vs , the objective of the voltage compensation is achieved without real power requirement by the DVR, even if the feeder resistance is non-zero. These aspects of the DVR will be discussed in the following section.

Figure 6.5 When the load is purely resistive

6.4.1

GENERAL CASE

In general, it is desired to maintain the magnitude of the load voltage equal to the source voltage i.e., 1.0 p.u. The voltage equation, in general, relating the source, load and DVR has been expressed in (6.3) and is given below. Vs

= V l + I Rs − (V f − jI Xs )

The above equation is illustrated using the phasor diagram description in Fig. 6.6 given below. Three cases of voltage compensation are discussed below.

Voltage Compensation Using Dynamic Voltage Restorer

283

Figure 6.6 Compensation using DVR

Case 1: When I Rs (= CD) < CD1 For this case, it is always possible to maintain the load voltage same as the source voltage, i.e., Vl = Vs . The DVR is expected to supply a wide range of reactive power to meet this condition. When I Rs is quite smaller than CD1 , the above condition can be met by supplying less reactive power from the DVR. For this condition, there are two solutions. Graphically, these solutions are represented by points A and B in Fig. 6.6. Case 2: When I Rs (= CD) > CD1 For this condition, it is not possible to meet Vl = Vs . This is shown by lines passing through points between D1 and D2 . This may take place due to the higher feeder resistance or high current, thus making the product of I Rs relatively large. Case 3: When I Rs (= CD) = CD1 This is a limiting case of compensation to obtain Vs = Vl . This condition is now satisfied at only one point when CD1 = I Rs . This is indicated by point D1 in Fig. 6.6. Now let us set the following objective for the load compensation. Vl

= Vs = V = 1.0 p.u.

(6.5)

From Fig. 6.6, OC = V cos φl = cos φl . Therefore, CD1 = OD1 − OC = V (1 − cos φl ) = (1 − cos φl ) p.u.. In order to meet the condition given by (6.5), the following must be satisfied. I Rs ≤ V (1 − cos φl )

(6.6)

284

Power Quality in Power Distribution Systems: Concepts and Applications

The above implies that V (1 − cos φl ) I V (1 − cos φl ) I≤ . Rs

Rs ≤ or

(6.7) (6.8)

Thus, it is observed that for a given power factor, the DVR characteristics can be obtained by varying Rs and computing I or vice-versa for a given value of φl , as described below. Let us consider three values Rs = 0.04 p.u., Rs = 0.1 p.u. and Rs = 0.4 p.u. For these values of feeder resistance, the line currents are expressed as follows using (6.8). I I I

= 25 (1 − cos φl ) p.u.

for Rs = 0.04 p.u.

= 2.5 (1 − cos φl ) p.u.

for Rs = 0.4 p.u.

= 10 (1 − cos φl ) p.u.

for Rs = 0.1 p.u.

The above currents are plotted as a function of load power factor and are shown in Fig. 6.7. Since I Rs = Vl (1 − cos φl ), when Rs increases, I has to decrease to make Vl (1 − cos φl ) to be a constant for a given power factor and load voltage. Thus, if the load requires more current than the permissible value, the DVR will not be able to regulate the load voltage at the nominal value, i.e., 1.0 p.u. However, we can regulate bus voltage less than 1.0 p.u. For regulating the load voltage less than 1.0 p.u., the current drawing capacity of the load increases.

Figure 6.7 DVR characteristics for different load power factor and feeder resistance

285

Voltage Compensation Using Dynamic Voltage Restorer

6.5

MATHEMATICAL DESCRIPTION TO COMPUTE DVR VOLTAGE

The previous section explains DVR characteristics and describes the feasibility of realizing DVR voltage graphically under different operating conditions. In this section, a feasible solution for the DVR voltage is presented with a mathematical description. This plays a significant role while implementing DVR on real time basis. Re-writing (6.1), V s +V f = V l + (Rs + jXs ) I Denoting, (Rs + jXs ) I = a2 + jb2 and V f = V f ∠V f = V f (a1 + jb1 ), the above equation can be written as following. Vs

= V l + (a2 + jb2 ) −V f

= V l + (a2 + jb2 ) −V f (a1 + jb1 ) = 1.0∠0◦ + (a2 + jb2 ) −V f (a1 + jb1 ) Since, source voltage and load voltage have to be maintained at nominal value, i.e., 1.0 p.u., therefore V s = Vs ∠δ = 1.0∠δ . Substituting this value of V s in above equation, we get,   V s = 1.0∠δ = cos δ + j sin δ = (1 + a2 ) −V f a1 + j(b2 −V f b1 )

Squaring and adding the real and imaginary parts from both the sides of the above equation, we get, (1 + a2 )2 +V f2 a21 − 2(1 + a2 ) a1 V f + b22 +V f2 b21 − 2 b1 b2 V f −Vs2 = 0

Since a21 +b21 = 1, therefore summation of underlines terms, V f2 (a21 +b21 ) = V f2 . Using this and rearranging above equation in the power of V f , with Vs = 1.0 p.u., we get the following. V f2 − 2 [(1 + a2 ) a1 + b1 b2 ] V f + (1 + a2 )2 + b22 − 1.0 = 0

(6.9)

The above equation gives two solutions for V f . These are equivalent to two points A and B shown in Fig. 6.6. However, the feasible value of the voltage is chosen on the basis of the rating of the DVR. In general, the load voltage may not always be 1.0∠0◦ p.u. Hence, the load voltage is expressed as V l = Vl ∠δl p.u. Substituting this value of the load voltage in (6.9), the following is obtained. Vs

= Vl ∠δl + a2 + j b2 −V f (a1 + j b1 ) = (Vl cos δl + a2 − a1 V f ) + j(Vl sin δl + b2 − b1 V f ) 



(6.10)

Denoting Vl cos δl +a2 = a2 and Vl sin δl +b2 = b2 , the above equation can be written as following. Vs cos δ + jVs sin δ

=





(a2 − a1 V f ) + j(b2 − b1 V f )

286

Power Quality in Power Distribution Systems: Concepts and Applications

Squaring and adding the real and imaginary terms in (6.10), the following is obtained. 





2



2

V f2 − 2(a1 a2 + b1 b2 )V f + a2 + b2 −Vs2 = 0

(6.11)

To solve DVR voltages for three-phase system, the nominal load voltages with their respective phase angles should be considered in (6.10). Therefore, V sk = (Vlk cos δlk + a2k − a1k V f k ) + j(Vlk sin δlk + b2k − b1k V f k )

(6.12) 

In the above equation, k denotes phases a, b, c. Denoting Vlk cos δlk + a2k = a2k and  Vlk sin δlk + b2k = b2k , the above equation can be written as following. Vsk cos δ + jVsk sin δ





= (a2k − a1k V f k ) + j(b2k − b1k V f k )

Squaring and adding the real and imaginary terms in (6.12), the following equation is obtained. 





2



2

V f2k − 2(a1k a2k + b1k b2k )V f k + a2k + b2k −Vsk2 = 0

(6.13)

Example 6.2. Consider a system with supply voltage 230 V = 1.0 p.u., 50 Hz as shown in Fig. 6.8. Consider feeder impedance as Zs = 0.05 + j0.3 p.u. and load impedance Zl = 0.5 + j0.3 p.u. (a) Compute the load voltage without DVR. (b) Compute the current and DVR voltage such that Vl = Vs . (c) Compute the effective source voltage including DVR. Explain the power flow in the circuit. (d) Compute the terminal voltage with DVR compensation.

Figure 6.8 A DVR compensated system

Voltage Compensation Using Dynamic Voltage Restorer

287

Solution: (a) When DVR is not connected The system parameters are given as follows. The supply voltage Vs = 1.0∠0◦ p.u., Zs = Rs + j Xs = 0.05 + j0.3 and Zl = Rl + j Xl = 0.5 + j0.3 p.u. The current in the circuit is given by, Is

1.0∠0◦ Vs = Zs + Zl 0.55 + j0.6 = 0.83 − j0.91 = 1.2286∠ − 47.49◦ p.u. = I = Il =

The load voltage is therefore given by, Vl

= Zl I s = 1.2286∠ − 47.49◦ × (0.5 + j0.3) = 0.7164∠ − 16.53◦ p.u.

Thus, we observe that the load voltage is 71% of the rated value. Due to a reduction in the load voltage, the load may not perform to the expected level. (b) When DVR is connected It is desired to maintain Vl = Vs by connecting the DVR. Taking V l as reference phasor, i.e., V l = 1.0∠0◦ , The line current is computed as below. I

=

1.0∠0 = 1.47 − j0.88 = 1.715∠ − 30.96◦ p.u. 0.5 + j0.3

Writing KVL for the circuit shown in Fig. 6.8, V s +V f = V l + (Rs + jXs ) I. The DVR voltage V f can be expressed as following. V f = V f ∠(∠I s + 90◦ ) The angle of V f is taken as (∠I s + 90◦ ) so that DVR does not exchange any real power with the system. Vf

= V f ∠(∠I s + 90◦ ) = V f ∠(−30.96◦ + 90◦ ) = V f ∠59.04◦ = V f (0.51 + j0.86) = V f (a1 + jb1 ) p.u.

The above equation implies that a1 = 0.51, b1 = 0.86. Let us now compute (Rs + jXs ) I. (Rs + jXs ) I

= (0.05 + j0.3) (1.47 − j0.88) = 0.3041∠80.54◦ × 1.715∠ − 30.96◦ = 0.5215∠49.58◦

= 0.3382 + j0.3971 p.u.

288

Power Quality in Power Distribution Systems: Concepts and Applications

This gives, a2 = 0.3382 and b2 = 0.3971. As discussed in previous section, the equation V s +V f = V l + (Rs + jXs ) I can be written in following form. V f2 − 2 [(1 + a2 ) a1 + b1 b2 ] V f + (1 + a2 )2 + b22 − 1.0 = 0.

Substituting a1 , b1 , a2 , b2 in the above equation, we get the following quadratic equation for the DVR. V f2 − 2.058V f + 0.9485 = 0

Solving the above equation, we get V f = 0.6969, 1.3611 p.u. as two values of the DVR voltage. These two values correspond to points A and B, respectively in Fig. 6.6. However, the feasible solution is V f = 0.6969 p.u., as it ensures less rating of the DVR. Therefore, Vf

= 0.6969∠59.04◦ = 0.3585 + j0.5976 p.u.

The source voltage can be computed using the following equation. Vs

= V l + (Rs + jXs ) I −V f . = 1.0∠0◦ + (0.05 + j0.3) 1.715∠ − 30.96◦ − 0.6969∠59.04◦ = 0.9797 − j0.2005 = 1.0∠ − 11.56◦ p.u.

(c) Effective source voltage It is seen that the magnitude of V s is 1.0 p.u. which satisfies the condition Vs = Vl . However, the angle of V s is ∠ − 11.56◦ , which implies that power is flowing from the load to the source. This is not true because the effective  source voltage is now V s = V s +V f . This is computed below. 

V s +V f = V s

= 0.9797 − j0.2005 + 0.3585 + j0.5976 = 1.3382 + j0.3971 = 1.3959∠16.52◦ p.u.

From the above, it is evident that the effective source voltage has a magnitude of 1.3959 p.u. and an angle of ∠16.52◦ which ensures that power flows from the source to the load. For this, the equivalent circuit is shown in Fig. 6.9. (d) Terminal voltage with DVR compensation The terminal voltage can also be computed as in the following. V t = V s − Zs I

= V l −V f

= 1.0∠0◦ − 0.6969∠59.04◦ = 0.8767∠ − 42.97◦ p.u.

This indicates that for the rated current flowing in the load, the terminal voltage is less than 1.0 p.u. After compensation, the load voltage is 1.0 p.u. as shown in Fig. 6.9.

289

Voltage Compensation Using Dynamic Voltage Restorer

Figure 6.9 A DVR compensated system

6.6

TRANSIENT OPERATION OF THE DVR

In the previous section, the operation of the DVR in the steady state was discussed with the assumption that complete system information is available. While implementing the DVR compensation scheme, the above discussed method should be implemented on a real-time basis. For the single-phase DVR operation, the following steps are required. 1. Define a reference quantity such as the load voltage vl (t), and then other quantities are synchronized to it. 2. The phase angle of the DVR voltage is obtained by computing the phase angle of the line current with respect to the reference quantity, as given below. ∠V f = ∠I l + 90◦ = ∠I s + 90◦ 3. Then DVR voltage is computed using (6.11), which is given below. 





2



2

V f2 − 2(a1 a2 + b1 b2 )V f + a2 + b2 −Vs2 = 0 4. DVR voltage v f is calculated using magnitude V f and phase angle obtained above in time domain form as given below. √ v f (t) = 2V f sin(ωt + ∠V f ) 5. The DVR voltage v f (t) is then synthesized using appropriate switching control of the voltage source inverter. The above method can be referred to as Type 1 control [15]. The method assumes that all circuit parameters are known along with the information of the source impedance. However, this may not be feasible in all circumstances. To solve this problem, Type 2 control is suggested. In Type 2 control, only local quantities are required to compute the DVR voltage. The method is described below.

290

Power Quality in Power Distribution Systems: Concepts and Applications

The terminal voltage, which is the local quantity to the DVR as shown in Fig. 6.1, can be expressed as follows. Vt

= V l −V f

= Vl ∠δl −V f (a1 + jb1 ) = Vl cos δl + jVl sin δl − a1 V f − jb1 V f =

(Vl cos δl − a1 V f ) + j(Vl sin δl − b1 V f )

(6.14)

Since, V t = Vt ∠δt = Vt cos δt + jVt sin δt , the above equation is written as following. Vt cos δt + jVt sin δt = (Vl cos δl − a1 V f ) + j(Vl sin δl − b1 V f )

(6.15)

Squaring and adding real and imaginary parts on both sides, we get the following. Vt2

= (Vl cos δl − a1 V f )2 + (Vl cos δl − b1 V f )2 = Vl2 +V f2 − 2(a1 cos δl + b1 sin δl )Vl V f

(6.16)

The above equation can be arranged in the powers of the DVR voltage as given below. V f2 − 2(a1 cos δl + b1 sin δl )Vl V f +Vl2 −Vt2 = 0

(6.17)

For V l as the reference voltage, δl = 0, the above equation is written as follows. V f2 − 2 a1 Vl V f +Vl2 −Vt2 = 0

(6.18)

We can see that the equation does not need the information of voltage drop across the feeder impedance, hence a2 and b2 are not present. The above equation is quadratic in V f and gives a positive value as a feasible solution for V f . If the roots are complex, it is impossible to compensate for the voltage without real power support. 6.6.1

OPERATION OF THREE PHASE DVR WITH UNBALANCE VOLTAGES WITHOUT HARMONICS

To implement DVR for the unbalanced three-phase system without harmonics, angles of DVR voltages are found by shifting current angles by 90◦ , i.e., ∠V f a

= ∠I a + 90◦

∠V f b

= ∠I b + 90◦

∠V f c

= ∠I c + 90

(6.19)



The magnitude of DVR voltage can be found using (6.9) and (6.17) for Type 1 and Type 2 control, respectively. Based on the above, the DVR voltages v f a , v f b , v f c can be expressed in the time domain as given below. √ vfa = 2V f a sin(ω t + ∠V f a ) √ vfb = 2V f b sin(ω t + ∠V f b ) (6.20) √ vfc = 2V f c sin(ω t + ∠V f c )

291

Voltage Compensation Using Dynamic Voltage Restorer

6.6.2

OPERATION OF THREE PHASE DVR WITH UNBALANCE VOLTAGES WITH HARMONICS

In the previous analysis, it was assumed that the supply voltages are unbalanced without harmonics. In this section, the operation of the DVR with harmonics will be discussed. The terminal voltages (vta , vtb and vtc ) are resolved into their fundamental voltages and the rest part, as given below. = vta1 + vta rest

vta vtb vtc

= vtb1 + vtb rest

(6.21)

= vtc1 + vtc rest

The angles of fundamental DVR voltages (∠V f a1 , ∠V f b1 , and ∠V f c1 ) are extracted as following. ∠V f a1 ∠V f b1 ∠V f c1

=

∠I a1 + 90◦

=

∠I b1 + 90◦

=

∠I c1 + 90

(6.22)



The angles of currents, i.e., ∠I a1 , ∠I b1 , ∠I c1 , are computed by extracting fundamental of the three-phase load currents using Fourier transform. The magnitudes of the fundamental DVR voltages (V f a1 ,V f b1 andV f c1 ) can be computed using equations (6.11) and (6.18) for Type 1 and Type 2 control, respectively. For example, using Type 2 control, the fundamental phase-a DVR voltage is computed as per the following equation. V f2a1 − 2 aa1 Vl V f a1 +Vl2 −Vta1 2 = 0

(6.23)

In (6.23), aa1 + jba1 = ∠V f a1 and Vta1 is fundamental phase-a terminal voltage as given above in (6.21). A similar expression can be written for phase-b and phase-c. This equation gives a solution only for the fundamental component of the DVR voltage. The rest of the DVR voltages consist of harmonics, which are equal and opposite to that of the terminal voltage harmonics, i.e., vta rest , vtb rest and vtc rest . Therefore, these can be given using the following equations. v f a rest v f b rest

= −vta rest

v f c rest

= −vtc rest

= −vtb rest

(6.24)

Thus, the total DVR voltage to be injected can be given as follows. vfa vfa vfa

=

v f a1 + v f a rest

=

v f b1 + v f b rest

=

v f c1 + v f c rest

(6.25)

In the above equation, v f a1 , v f b1 , and v f c1 are constructed using (6.20). Once v f a , v f b , and v f c are known, these voltages are synthesized using a suitable power electronic circuit. This is discussed in the following section.

292

6.7

Power Quality in Power Distribution Systems: Concepts and Applications

REALIZATION OF DVR VOLTAGE USING VOLTAGE SOURCE INVERTER

In the previous section, a reference voltage of DVR was extracted using discussed control algorithms. This DVR voltage, however should be realized at the power level. This is achieved with the help of a power electronic converter which is also known as a voltage source inverter. Various components of the DVR are shown in detail in Fig. 6.10. The transformer injects the required voltage in series with the line to maintain the load bus voltage at the nominal value. The transformer not only reduces the voltage requirement but also provides isolation between the inverters and the line. The filter components of the DVR, such as external inductance (Lt ), which also includes the leakage of the transformer on the primary side and ac filter capacitor on the secondary side, play a significant role in the performance of the DVR [15]. The

Figure 6.10 A DVR compensated system

same dc link can be extended to other phases as shown in Fig. 6.10. The single-phase equivalent of the DVR is shown in Fig. 6.11. In Figs. 6.10 and 6.11, vinv denotes

Figure 6.11 Equivalent circuit of the DVR

the switched voltage generated at the inverter output terminals. The resistance, Rt , models the switching losses of the inverter and the copper loss of the connected transformer and filter inductance. The voltage source inverter (VSI) is operated in a

293

Voltage Compensation Using Dynamic Voltage Restorer

switching band voltage control mode to track the reference voltages generated using control logic as discussed below. Let V f∗ be the reference voltage of a phase that DVR needs to inject in series with the line with the help of the VSI explained above. We form a voltage hysteresis band of ±h over the reference value. Thus, the upper and lower limits within which the DVR has to track the voltage can be given as follows. v f up = v∗f + h v f dn = v∗f − h

(6.26)

The following switching logic is used to synthesize the reference DVR voltage. If v f ≥ v f up

S1 , S2 OFF, and S3 , S4 ON (‘-1’ state) else if v f ≤ v f dn

S1 , S2 ON, and S3 , S4 OFF (‘+1’ state)

end It is to be noted that switches status S1 − S2 ON and S3 − S4 OFF is denoted by ‘+1’ state and it gives vinv = +Vdc . The switches status S1 − S2 OFF and S3 − S4 ON corresponds to ‘-1’ state providing vinv = −Vdc as shown in Fig. 6.10. The above switching logic is very basic and has a scope to be refined. For example ‘0’ state of the switches of the VSI as shown in Fig. 6.10, can also be used to have smooth switching and to minimize switching losses. In the zero state, vinv = 0 and refers switches status as S1 D3 or S2 D4 for positive inverter current (iinv > 0). Similarly, for negative inverter current (iinv < 0), ‘0’ state is obtained through S3 D1 or S4 D2 . With the addition of ‘0’ state, the modified switching logic becomes as follows. If v∗f ≥ 0

if v f ≥ v f up

‘0’ state

else if v f ≤ v f dn ‘+1’ state

end else if v∗f < 0 if v f ≥ v f up

‘-1’ state else if v f ≤ v f dn ‘0’ state

end end

294

Power Quality in Power Distribution Systems: Concepts and Applications

In order to improve the switching performance, one more term is added in the above equation based on the feedback of filter capacitor current, i f ac . v f up = v∗f + h + α i f ac v f dn = v∗f − h + α i f ac

(6.27)

Where α is a proportional gain given to smoothen and stabilize the switching performance of the VSI [16]. The dimension of α is Ohm and is equivalent to virtual resistance, whose effect is to damp out and smoothen the DVR voltage trajectory resulting from the switching of the inverter [20]. The value of the hysteresis band (h) should be chosen in such a way that it limits switching frequency within the specified maximum value. This kind of voltage control using VSI is called as switching band control. The actual DVR voltage is compared with these upper and lower bands of the voltage (v f up , v f dn ), and accordingly switching commands to the power switches are generated. The switching control logic is described in Table 6.1. To minimize the switching frequency losses of the VSI, three level logic has been used. For this, an additional check of the polarity of the reference voltage has been taken into consideration. Based on this switching status, the inverter supplies +Vdc , 0, and −Vdc levels of voltage corresponding to the 1, 0, and -1 given in the table in order to synthesize the reference DVR voltage.

Table 6.1 Three level switching logic for the VSI Conditions

Switching value

v∗f ≥ 0

v f > v f up

0

v f < v f dn

1

v f up